Livestock Research for Rural Development 28 (1) 2016 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Socio-economic determinants of milk production in Bangladesh: an implication on on-farm water use

M N Sultana, M M Uddin and K J Peters1

Department of Animal Nutrition, Bangladesh Agricultural University, Mymensingh-2202, Bangladesh
1 ADT-Institute for Agricultural and Horticultural Sciences, LeWi-Faculty, Humboldt University of Berlin, Germany


This study envisaged to understand the socio-economic factors that affect milk production and on-farm blue water use (OBWU). The cross sectional data sets collected from 220 sample dairy farms were modeled by using two separate regression models. Firstly, a general linear model (GLM) was used to explore the factors that influence milk production. Secondly, OBWU (categorized as low, medium and high) was regressed by using a multinomial logit (MNL) model to identify socio-economic determinants of OBWU.

The GLM results revealed that age of farm owner, off-farm income and training had a negative influence, while experience in farming activities have positive impact on milk production. The MNL regression results revealed that gender of the farm owner (men), education, experience, milk yield, and herd size had a significant positive influence on OBWU, and the increasing age of farm owner and off-farm income had a negative significant impact on OBWU level.

It is concluded that socio-economic factors affecting milk production and OBWU are highly linked with adoption of efficient management decision tools that will guide the farmers to allocate water resources effectively for increasing milk production.

Key words: blue water, farmer experience, socio-economic factors


The livestock industry is charged with providing sufficient animal products to meet the market demand while it needs to improve the environmental perspective of animal production (Capper and Bauman 2013). The livestock production in Bangladesh (BD) is undergoing rapid socio-economic and production system changes that raise the question on the ability to maintain the current milk production to meet the growing demand for the welfare of consumers. In addition, climate change projections indicate an increased likelihood of droughts and uneven distribution of rainfall leading to increasing phenomenon of water scarcity and temporal as well as spatial availability. Therefore, the water problem in BD is an alarming issue owing to growing demands, climate change and increasing conflict between current practice and alternative options for water use (Rahman and Parvin 2009; Chowdhury 2010). Decreased water availability, thus, is a risk factor to food security (Rahman and Parvin 2009) and this would heavily affect the livelihoods of farmers and hamper the development of the country (Karim et al 2010).

Bangladesh has both economic and physical water scarcity (Iwmi 2008; Gaufichon et al 2010) along with an average water stress index (WSI) of 0.499 (WSI range: 0 to 1, Pfister et al 2009). Economic water scarcity (EWS): areas having adequate renewable resources with less than 25% of water from rivers withdrawn for human purposes, but needing to make significant improvements in existing water infrastructure to make such resources available for use, are considered EWS. Physical water scarcity (PWS): when more than 75% of river flows are withdrawn for agriculture, industry and domestic purposes is called PWSd. On the other hand, there is an increasing demand to reach food security which has led to an increase of blue water-irrigated agriculture accounting for 80-90% of the total water use (GOB 2010; Rahman and Parvin 2009).

Livestock production, particularly, dairying is the source of livelihood for about 50-70% of people in rural Bangladesh (Ser-Od et al 2008) and dairy products are an important protein supplement to rice. Therefore, demand expressed as consumption of milk and milk products is increasing at a faster rate (5.0%) than the production of milk from cows, buffaloes, sheep and goats (4%) (Hemme et al 2012). This instigates to expand dairying much faster than before but the water scarcity poses an increasing challenge due to possible inter-linkages and competition between the water and the food supply system for human beings (Strzepek and Boehlert 2010). This issue is linked to the significant problems related to human health, livelihood, social or political stability and environmental sustainability (United Nations Global Compact 2012).

Translating the national water scarcity to farm level, it is evident that mismanagement in livestock production is one of the major contributors to water scarcity and environmental degradation in the integrated livestock production system (i.e. crop and dairy) (Tegegne 2012). The factors that provoke mismanagement in dairy production are diverse from the point of view of socio-economic development, climate, water availability, infrastructure, and the social, ecological, political pressure on water resources (Iglesias et al 2007). Water scarcity is a serious problem during dry seasons, because of low and uneven distribution of rainfall, insufficient storage options and drying-up of several canals and rivers which otherwise could be used for agriculture production (Chowdhury 2010). The availability of water in Bangladesh is characterized by two opposite magnitude between dry and monsoon vis-à-vis, water scarcity and flood, example: Very high scarcity during dry hot season (March to June), moderate scarcity during dry wet season (November to February) but over flow of water, i.e. flood during monsoon (July to September). Thus, water scarcity in BD is defined as scarcity from lack of water during dry season and scarcity from over flooded water during monsoon. During flood, water is everywhere but potable water is scarce for animal as well as human (Chowdhury 2010).

Water use pattern plays a pivotal role for regional water scarcity (Ridoutt et al 2010). To combat the anticipated water scarcity (Quasem et al 2010), there is a need for strategies to improve livestock water productivity and efficient water use management (Van Breugel et al 2010). But the efficient management in milk production is constrained by socio-economic factors (Brian et al 2009; Quasem et al 2010; Uddin et al 2012). Understandings of farm specific socio-economic factors have direct relevance for management decisions on water resource allocation that would help to increase milk production, support economic development and sustain resilience in poor rural communities. The current understanding of human values and the way to incorporate socio-economic factors in the decision making process is limited (Lockwood 1999). Concurrently, intensification of milk production is sought to increase all over the world (Alvarez et al 2008) as responding to favorable market demands. But this is threatening the sustainability of farming systems and leads to expanded inefficient utilization of natural resources, particularly freshwater consumption at farm level (Sultana et al 2014).

Therefore, this research envisages testing empirically how the socio-economic factors influence milk production and direct on-farm blue water use (OBWU). OBWU is defined as surface or ground water and it is mainly appropriated into dairy products life cycle, i.e., irrigation water in feed and forage production, and drinking and servicing water in farming activities. In this study, water requirement for feed and fodder production was excluded as >90% dry matter of the daily ration consist of rice straw by-product (Sultana et al 2015) which has a water use about zero because the water use for growing crop is mainly attributed to the main products, not the low-value crop by-products (Mekonnen and Hoekstra 2010).

On the other hand, the direct consumption of blue water has an impact on the contribution of local water scarcity while green water (i.e. soil water used in evapotranspiration process by plants) is not like blue water which does not limit the availability blue water or environmental flows and has no same impact in a local hydrological system (Berger and Finkbeiner 2010; Ridoutt et al 2010; Doreau et al 2012). Consequently, OBWU for drinking and servicing in farming can represent a significant amount of high-quality water in competition with human water consumption during the dry period and even in monsoon rains flood. In some cases, dairy cattle watering can represent a significant amount of additional cost for the farm, if supplied by potable public water (Khelil-Arfa et al 2012). The discourse has turned its focus to demand effective water management. Understanding socio-economic factors that influence water use for increased milk production is crucial when devising effective water management strategies at farm level. Therefore, this study envisaged to understand the socio-economic factors. In order to address this research objective, the following two research questions are addressed in this study: i) how and to what extent do the socio-economic factors influence milk production; ii) what are their implications on direct OBWU?

Materials and methods

Study areas

Districts in the north-west (NW) and north-eastern (NE) parts were selected to provide the farm data base. The altitude varies between 36 m, and 10 m (Sumiko 1993). Three districts were selected from these two regions. Two districts were belonging to NW region (i.e. Dinajpur and Sirajgonj) and one was in NE region (Kishorgonj). Figure 1 represents the study locations with agro-ecological condition, rainfall and temperature.

Figure 1. Milk density and environmental condition based on temperature(°C) and rainfall (liters/ha/year)

The NW is bounded by the Brahmaputra-Jamuna and Ganges - two major rivers in South Asia. This region has 34,600 km2 in area and the land is dominated by high (36 m) and medium (16 m) altitude of land (Sumiko 1993; BBS 2007). Agriculture, particularly dairying, is the major activity in the NW region but water is the liming factor to promote dairying and hampers the development. A shallow tube well (STW) irrigation pumping technique is the pre-dominant source for water due to which seasonal water table decline is widespread within the region (Halcrow 2001). Apart from that, the water availability in this region is also dependent on the actual water flow in the Ganges River which highly fluctuates due to the construction of the Farakka Barrage (which is a Dam constructed by India upstream).

On the other hand, the NE zone is situated under the ‘Old Brahmaputra’ and it has 7230 km2 in area and is dominated mainly by low (10 m) land altitude (48%). This region belongs to the ‘Haor’ area (literally the area “surrounded by water all the year round”). This region belongs to a wet area and hence, has a little exploitable shallow ground water and dominantly uses surface water resources during dry season. Irrigation mostly depends on low lift pumps than shallow tube wells (Halcrow 2001). These are the major factors that differentiate this area from the NW region. The key selection reasons are: 1) these regions are considered to have potential for high performance dairy and produces already >40% of the total national milk (Hemme et al 2004); 2) several dairy development programs and interventions have been implemented in this region (Zaedi et al 2009); 3) the sample dairy farms were expected to represent the typical dairy farming systems and the results of this sample population are expected to be transferable to other dairy regions; 4) agriculture, particularly dairying, is the major activity and the return from dairying to the total household income ranges from 45 to 94% (Uddin et al 2013); 5) water access, sources of water supply and water availability varies among farms, being the major liming factor to promote dairy during the dry winter season (insufficient rainfall) and wet monsoon season with its “water over flow (i.e. flood)” (Shirazi et al 2011).

Sampling procedure and data collection

In order to make valid inferences and to represent the overall milk production systems and the consequences of different levels of water use on milk production, it is necessary to collect data from all the possible observations of interest. A simple random sampling technique is appropriate in such case that helps to avoid biasness in selection of dairy farms (Quinn and Keough 2002) because in this technique, all dairy farms have the same probability of being selected. Hence, within a radius of 500 m from the center of each study village 3-5 sample dairy farms were selected randomly. In total 44 villages (22 villages per region) were selected based on the facts that dairy is the major activities for more than 40% of the village population.

A pre-designed and pre-tested questionnaire was used to collect the farm data by using face-to-face interviews. In total 220 dairy farms were interviewed. The data collection was done on the following broad subjects: 1) Milk production, 2) OBWU, such as water for drinking, feed mixing plus cleaning roadside grass and servicing (i.e. cleaning dairy shed, utensils, and dairy animals themselves etc; 3) Sources of water supply; and 4) Questions on socioeconomic factors, and farm and farmer characteristics. The description of the variables and their descriptive statistics are stated in Table 1 and Table 2, respectively.

Table 1. Definition and description of the variables Socio-economic factors analyzed

Variable code



Age of the farm owner in years (≤25 years = 1; 2 = 26-35 years = 2; 36-45 years = 3; 46-55 years = 4; 56-65 years = 1;≥ 66 years = 6)


Level of formal education of the farm owner (No = 0, Primary = 1, Secondary = 2, College = 3, University = 4)


Experience is indication of number of years involved in dairying by farm owners (≤2 years = 1; 3-5 years = 2; 6-10 years = 3; 11-15 years = 4; ≥15 years = 5)


Number of family members living on the dairy farms


Gender of the farm owner (Male = 1; Female = 2)


The size of the dairy herd (i.e. the number of dairy animals per farm)


The amount of milk produced (kg/cow/day)


The total water use per cow per day in liters (1=low, 2=medium, 3= high) and this OBWU was further categorized into three different levels: i) low ranging from 1 to 200; ii) the medium from 201 to 399 and iii) high more than 400.


Importance of off-farm income (No = 0; Yes =1)


Training (i.e. feeding management and other related dairy activities) received by farm owner during last two years (No =0, Yes = 1)

Table 2. Descriptive statistics of socio-economic factors and milk yield used for modeling



Std. dev.





















































*The outlier large herds (128 animals) were excluded during further analysis.

Statistical model

For the first research question, a General Linear Model (GLM) [1] regression in which milk yield was used as dependent (Y = response) variable and socio-economic factors as well as the OBWU were used as independent (X = predictor) variables which are stated below:

where β0 = constant, β1 = coefficient, Xi = the X-values were fixed that represent the socio-economic factors and total OBWU, i = error, Yi = the Y-value for each observation is sampled from a population of all possible Y-values.

For addressing the second question [2], a multi-nomial logit regression model (MNL) was applied in order to analyze the implication of socio-economic factors on total OBWU. Maddala (1983) reported that this method has superiority over any other categorical variable estimation technique because in this model, explanatory variables signify effects on qualitative variables. Although a binary choice logit model is widely used (Kim et al 2005), a discrete choice logit model, for example, a MNL model, is applicable in our study because the dependent variable is OBWU that contains more than two categorical variables which have no natural ordering (STATACORP 2009). Therefore, this study followed the method principles which can be stated as:

where m represents different levels of OBWU. The MNL model considers three unordered outcomes: low (base category), medium and high. This unordered categorical property of the dependent variable distinguishes the MNL from traditional regression models where the dependent variable is continuous, and from ordered Logit models where the dependent variable r is ordered, and from Logit models which are appropriate for binary (0,1) outcomes. To estimate the relationship between the level of milk yield intensity and OBWU using a chi-square test was positive and significant at 1% level (chi-square value = 15.2, with p = 0.004) and this justifies the validity for using OBWU as a response variable.

Estimation of the model

First cross-sectional data on milk production (kg/cow/day) was used as response variable (Y) and all socio-economic variables and total OBWU were modeled as predicted variables which are shown below:

On the other hand, the empirical MNL model estimated can be expressed in general terms as:

where Yi represents the probability of using different levels of OBWU, β’s and γ’s are the unknown parameters.


Milk yield and water use in the study area

The number of farms belonging to the different levels of OBWU and milk yield is depicted in Figure 2. Among 220 dairy farmers 62 % (135 dairy farmers) use a low (up to 200 L/day), 25 % (55 farmers) use a medium level (201 to 399 L/day), and only 29 dairy farmers (13%) use a high (>400 L/day) level of OBWU. The average milk yield of the low, medium and high OBWU group was 7, 9 and 9 kg/cow/day, respectively.

Figure 2. Number of farms belonging to various levels of OBWU and average milk yield of cows

The contribution of servicing, drinking and water for feed cleaning plus mixing is depicted in Figure 3.

Figure 3. Contribution of water for servicing, drinking and feed cleaning plus mixing to total OBWU

The highest amount of OBWU (L/cow/day) is the servicing water (60 L) followed by drinking water (55 L) and water for feed cleaning plus mixing (18 L). In terms of proportion to total OBWU, 45 % is used for servicing water, 41% for drinking and 14% for feed cleaning plus mixing.

The sources of OBWU on the dairy farms are related to a management pattern within the water supply systems. The study showed that sample dairy farmers use different sources to procure the daily needs of water for the cows as shown in Figure 4. The majority of the farmers (63%) use ground water by using either hand pump wells (45%) or piped water supply powered by electric motor (18%), surface water is used by about 2% of the farmers (ponds/river/canal), and iii) 35% use ground and surface water.

Figure 4. Sources of on-farm blue water supply for milk production
Socio-economic factors and OBWU affecting milk production

The regression model was used to test the hypothesis that the level of milk yield and socio-economic factors are related. The regression model was estimated by “Ordinary Least Square” (OLS) depicted in Table 3. The R-square of 0.35 indicates the limited fit of the model explaining only 35% of the variation in milk yield by the independent variables included. The p value of p=0.00 also indicates that the model rejects the null hypothesis that all coefficients except intercept are zero. The coefficients from GLM are also possible to interpret in a linear fashion because of the normal distribution of the milk yield data. The coefficient for age (AGE) is negative and significant at 10% level, indicating that milk yield decreases with increasing age of the farmers. Access to training of the farm owner (TRA) significantly and negatively (at 1% level) affects milk yield. This result was unexpected because training is expected to increase the managerial skills of the farmers and should enhance skills for increasing milk yield. However, the effect of level of experience of the farmers (EXP) on milk production was positive and significant (p<0.05) because more experienced farmers can properly manage resources for higher milk yield than non-experienced farmers.

Table 3. Factors affecting milk production based on the parameter estimates of socio-economic variables and water use from GLM



Std. Err.

p value





































Off-farm income (OFINCOME) showed a negative effect on milk yield at 5% level of significance. This result was also in line with the expected hypothesis since the farms that depend on other sectors for ensuring the family expenditure might invest less managerial skills on dairy farming. It is more interesting that a higher education of farmers (EDU) seems to be unrelated to milk yield which is again somewhat unexpected. Apart from the socio-economic factor, OBWU (p<0.01) is positively related with milk yield. The reasons might be due to animals is not getting enough drinking water as well as the farms do not use sufficient water for feed mixing (e.g. water used for feed mixing).

Socio-economic factors affecting OBWU

The socio-economic factors influencing levels of OBWU using the MNL model are presented in Table 4. The model fits well because it rejects the null hypothesis that all coefficients except the intercept are zero. The likelihood ratio with chi-square of 116.96 with a p-value <0.0001 tells us that the model as a whole fits significantly better than an empty model (i.e. a model with no predictors).

Table 4. Factors affecting on-farm blue water use based on parameter estimates of socio-economic variables and milk yield from MNLa



Std. Err.

p value









































.a Low level of water use is the base outcome.

The percentage of concordant responses used to determine the predictive ability of the model is 42%, which signifies that the model explains 42% of the variation in OBWU, indicating a moderate fit. The formulation of the MNL model requires a base outcome (omitted category) which in this case is the low level of OBWU (Table 4).

The results of MNL showed the probabilities of different levels of OBWU as a function of several socio-economic factors. The MNL assumes that data are case specific; that is, each level of water use (low, medium and high) variable has a single value for each case. The MNL also assumes that the dependent variable cannot be perfectly predicted from the independent variables for any case (Green 2008). However, the key strength of this model is that as with other types of regression, there is no need for the independent variables to be statistically independent from each other (unlike, for example, in a Naive Bayes classifier); however, co-linearity is low, as it becomes difficult to differentiate between the impact of several variables if they are highly correlated. In this regard, our data set fits well to the MNL because the independent variables of socio-economic factors attributed a relatively low co-linearity, mainly because these factors are very much identical to the specific farm and have substantial variation among each other. Among 9 variables, seven variables such as age, sex, education, experience, off-farm income, herd size and milk yield significantly influence the level of OBWU although their level of significance and direction of impact are different. The most important findings of the MNL model are: 1) age and off-farm income are associated with a lower level of OBWU (p<0.05); 2). Gender of the farm owner (male) is associated with a positive impact indicating a higher likelihood of using a high level OBWU; 3) Education and experience are associated with higher level of OBWU at 5 and 10% level of significance; 4) Farm economy of scale (herd size) and milk yield are positively and significantly associated with level of OBWU at 5% and milk yield at 10% level of significance, respectively. This might be explained that farmers with higher herd size produce more milk/cow and consequently required more OBWU.


Results found that service water use has the highest contribution to total water use (Figure 3). The reason is that in summer, cattle need two times bathing. However, besides servicing (including bathing of animals) processes, losses from leaky buckets, tubes and watering places may cause a huge loss of water. The ground water use via well supply system is the major contribution of water supply (Figure 4) which is supported by Islam et al (2010). A number of socio-economic factors have been identified in this study which affects both milk production and OBWU as complied in Table 3 and 4. Milk production is affected by age, experience, off-farm income and training of farm owners (Table 3). Age has a negative relation to water use and milk yield which derives the hypothesis that the increase in age is associated with a decrease in OBWU and milk yield. This might be explained by the facts that older farmers are less likely to seek changes and challenges (i.e. they are reluctant toward the adoption of advanced technology) but they are more likely to be embedded in social networks (Schwartz 2003). In contrast, young dairy farmers tend to attain higher dairy outputs. This result is consistent with Gale (1994), who suggests that producers of older age “may reduce farm work load as their health declines or to accommodate reduced income needs”. The probability of disinvestment behavior is reflected through the negative relation between milk production and age of owner (Gale 2003).

Experiences in dairy farming show a positive and significant influence on both OBWU and milk yield. This indicates that gaining experience might enhance efficient management decision on water use and dairy management. This is possible through several avenues: i) receiving formal education and training, ii) interaction with an informal network, and iii) learning by doing. A significant and negative impact of training observed in this study is against the hypothesis. This could possibly be due to the fact that the majority of the farmers (86%) have no access to training (Table 2) and the existing training services are mostly focused on general farm operation or even to overall agricultural practices rather to farm specific resource use to dairying. However, the low access to training (14%) found in this study is also in line with results obtained by Jabbar et al (2005) and Uddin et al (2011) who found relatively poor access to training services provided by different organizations. This necessitates increasing the farmer’s access to training facilities that focus solely on milk production including water resource management knowledge.

Gender and education of the farm owner are found to have a significant influence on OBWU while no significant impact is observed for milk production. Men are predominantly household heads in charge of management decision in the dairy farm (Zaedi et al 2009). Thus, the gender (male)-driven management approach influences milk production and OBWU. This might be explained by the facts that male farmers are likely to be more skilled and able to decide in farm management efficiently in Bangladesh, while females still linked to the traditional role and tend to attribute more importance to benevolence and less importance to decision power (Schwartz and Rubel 2005).

The level of education is positively affecting the OBWU mainly due to the fact that educated farmers are always considered as progressive farmers who might adopt innovative management decision on resources use in milk production. The farmers who engage in off-farm activities are eventually less efficient in water use in dairying. The negative influence of off-farm income on OBWU might be interpreted in the way that farmers who depend relatively more on off-farm income to maintain their family find less time to invest on own resource uses (Hemme and Uddin 2009).

A farm having higher milk yield and herd size needs substantially more water per day. The higher the milk yield (cow/day), the higher the water required due to their maintenance and productivity requirements (Thomas 2011). For example, drinking water is directly related to Dry Matter Intake (DMI) which again relates to milk yield. A restriction of drinking water by 40% is associated with a 16% decrease in the DMI and a concomitant decrease in milk yield by 16% (Little, 1976). A previous study by Sultana et al (2014) stated that farmers with larger herds were also more likely to purchase more concentrate and dry feed composed of crop residues (rice straw, wheat barn, rice polish etc.) as the main component added to green grass to the ration (10%) and therefore, require more water for feed cleaning plus mixing and drinking to increase digestibility. Therefore, farmers with larger herds were also more likely to have more purchased dry feed which would require more water for mixing per farm.

This study confirms the critical importance of increasing efficiency in water use from the socio-economic and farming environment point of view. In this regards, improving water use efficiency in mixed crop-livestock systems is important both for people’s livelihoods and environmental resilience to reduce the pressure on this scarce resource. Devising effective water management considering the socio-economic factors is one promising way to increase milk production while reduce pressure on water resources. Nevertheless, the results obtained from this study have the following implications for efficient OBWU management decisions: i) within existing farming circumstances, it is necessary to expose dairy farmers to training in including water importance in milk production in order to disseminate and share the knowledge and experience among dairy farmers; ii) the farmer’s educational status on dairy farming needs to be improved, which can be done by strengthening government extension service; iii) there might be the need to provide incentives to motivate relatively older farmers to become aware on the negative consequences of reluctant management on milk production and water use.

Therefore, socio-economic impacts of water management actions are an emerging demand for improvement of water use for future food production through sustainable farming. While milk is the primary output that contributes 72 - 82% to the total farm income in BD (Uddin et al 2010). Zaedi et al (2009) and Hall et al (2012) found that OBWU has influence in increasing milk production within integrated dairy production system, thus, it has the potential to foster livelihoods through higher dairy income compared to the national level household income.



The authors highly acknowledge the International Foundation for Science (IFS), Sweden for providing the grant to conduct the Field Study in Bangladesh (Grant Reference B/5117-1). We sincerely thank Bradley G Ridoutt (CSIRO, Sustainable Agriculture Flagship) who provided critical comments and expert advice on water in farming practice.

[1] Standard linear regression models with standard estimation techniques make a number of assumptions about the predictor variables, the response variables and their relationship. To avoid this, this study utilized the GLM which is the extension of linear model but have advantages of having assumptions on normality and variance are relaxed.

[2] For details of both the unconditional and the conditional logit model see, for example, GREEN (2008).


Alvarez A J, Corral D, Solis D and Perez J A 2008 Does intensification improve the economic efficiency of dairy farms? Journal of Dairy Science, 91, 3693-3698.

Berger M and Finkbeiner M 2010 Water footprinting: How to assess water use in life cycle assessment? Sustainability 2, 919-944.

Brian O, Mpairwe D, Mutetika D, Bashasa B, Kiwuwa G and Peden D 2009 Socioeconomic factors affecting livestock water productivity in rainfed pastoral production systems.

Capper J L and Bauman D E 2013 The role of productivity in improving the environmental sustainability of ruminant production Systems. Annual Review of Animal Biosciences 1, 469-489.

Chowdhury N T 2010 Water management in Bangladesh: an analytical review. Water policy 12, 32-51.

Doreau M, Corson M and Wiedemann S G 2012 Water use by livestock: A global perspective for a regional issue?

Gale H F J R. 1994 Longitudinal analysis of farm size over the farmer’s life cycle. Review of Agricultural Economics, 16, 113–123.

Gale H F J R. 2003 Age-specific patterns of exit and entry in U.S. farming, 1978–1997. Review of Agricultural Economics, 25, 168-186.

Gaufichon L, Prioul J L and Bachelier B 2010 What are the prospects for genetic improvement in drought-tolerant crop plants? FARM Foundation, c/o Crédit Agricole S.A., 91 –93 Boulevards Pasteur, 75710 Paris Cedex 15, France.

GOB (Government of Bangladesh) 2010 Survey and monitoring project for development of minor irrigation. Bangladesh agricultural development corporation (BADC).

Green W H 2008 Econometric analysis, 6th ed., Prentice Hall: Englewood Cliffs.

Hall D C, Alam M G S and Raha S K 2012 Improving dairy production in Bangladesh: Application of integrated agriculture and eco-health concepts. International journal of livestock production 3 (3), 29-35.

Hemme et al 2012 Dairy Report 2012. International Farm Comparison Network. IFCN Dairy Research Center, Schauenburger Str. 116, 24118 Kiel, Germany. Available at:

Hemme T, Garcia O and Khan A R 2004 A Review of Milk Production in Bangladesh with particular Emphasis on small-scale producers. FAO-Pro-Poor Livestock Policy Initiatives working paper No. 7

Hemme T and Uddin M M 2009 Dairy Policy impacts on Bangladesh and EU 15 dairy farmers’ livelihoods: Dairy Case study. International Farm Comparison Network (IFCN) Dairy Research Center, Kiel, Germany. Available at: study-EU--BD-01-10-2010---website-version.pdf .(Accessed on January 3, 2010).

Iglesias A, Garrote L, Flores T and Moneo M 2006 Challenges to manage the Risk of water scarcity and climate change in the Mediterranean. Water Resource Management 21, 775-788.

Islam K M A Uddin M M, Sultana M N. Assaduzzaman M and Islam M N 2010 Distribution pattern and management practices of cross bred dairy cows in cooperative dairy production system in Bangladesh. Livestock Research for Rural Development 22 (12).

IWMI (International Water Management Institute) 2008 “Areas of physical and economic water scarcity”. UNEP/GRID-Arendal Maps and Graphics Library.

Jabbar MA, Islam S M E, Delgado C Ehui S, Akanda M A I, Khan M I and Kamruzzaman M M 2005 Policy and Scale factors influencing efficiency in dairy and poultry producers in Bangladesh. Joint working paper by ILRI/SLP/BSMRAU. Available at (Accessed on December, 2013).

Karim Z, Huque LK S, Hussain M G, Ali Z and Hussain M 2010 Growth and development potential of livestock and fisheries in Bangladesh. Bangladesh Food security and Investment Forum.

Khelil-Arfa H A, Boudon A, Maxin G and Faverdin P 2012 Prediction of water intake and excretion flows in Holstein dairy cows under thermo neutral conditions. Animal.doi:10.1017/S175173111200047X.

Kim S, Gillespie P J and Paude P J 2005 The effect of socio-economic factors on the adoption of best management practices in beef cattle production. Journal of Soil and Water Conservation (60), 111-120.

Little W, Sansom B F, Manston R.and Allen W M 1976 Effects of restricting the water intake of dairy cows upon their milk, body weight and blood composition. Animal science 22, 329-339

Lockwood M 1999 Preference structures, property rights, and paired comparisons. Environmental and Resource Economics 13, 107-122.

Maddala, G 1983 Limited Dependent and Qualitative Variables in Econometrics. Cambridge University Press, New York, p. 457.

Mekonnen M M And HoekstraA Y 2010 The Green, Blue And Grey Water Footprint Of Farm Animals And Animal Products. Vol. 1: Main Report. Value Of Water Research Report Series No.48.

Pfister S, Koehker A and Hellweg S 2009 Assessing the environmental impacts of freshwater consumption in LCA. Environmental Science and Technology, 43 (11), 4098-4104.

Quasem M A and Yasmin F 2010 Agricultural research priority: Vision-2030 and beyond. Bangladesh agricultural research council, Farmagte, Dhaka.

Quinn G and Keough M 2002 Experimental Design and Data Analysis for Biologists, Cambridge University Press, New York, USA.

Ridoutt B G and Pfister S 2010 A revised water footprinting to make transferent the impacts of consumption and production on global freshwater scarcity. Global Environmetal Change 20: 113-120.

Rahman M W and Parvin L 2009 Impact of irrigation on food security in Bangladesh for the past three decades. Journal of water resource and protection.doi:10.4236/jwarp.2009.13027

Schwartz S H 2003 A proposal for measuring value orientations across nations, in:European Social Survey: The questionnaire development package of the European Social Survey, Chapter 7, 259–319.

Schwartz S H and Rubel T 2005 Sex differences in value priorities: Cross-cultural and multi-method studies, Journal of Personality and Social Psychology, 89 (6): 1010–1028.

Ser-Od T M M Hussain and Dugdil B 2008 Improved Market Access and Smallholder Dairy Farmer Participation for Sustainable Dairy Development. Animal Production and Health Commission for Asia and the Pacific and Food and Agriculture Organization (APHCA-FAO) Project report presented in Asia-Pacific Smallholder Dairy Strategy Workshop, Chingmai, Thailand, 25-29 February 2008.

Shirazi S M Z, Ismail I, Shatirah A, Sholichin M and Islam M A 2011 Climatic parameters and net irrigation requirement of crops. International Journal of the Physical Sciences Vol. 6(1), pp. 15-26. DOI: 10.5897/IJPS10.582

STATACORP 2009 Stata 11 Base Reference Manual. College Station, TX: Stata Press.

Trzepek K and Boehlert B 2010 Competition for water for the food system. Philosophical of Transactions of the Royal Society, B 365 (1554), 2927-2940.

Sultana M N, Uddin M M, Ridoutt B G, Hemme T and Peters K J 2014 Comparison of water use in global milk production for different typical farms. Agricultural Systems129: 9-21

Sultana M N, Ridoutt B G, Uddin M M, Hemme T and Peters K J 2015 Benchmarking consumptive water use of bovine milk production systems for 60 geographical regions: An implication for Global Food Security. Global Food Security 4: 56-68

Sumiko K 1993 Geomorphological features of northwestern Bangladesh and some problems on flood mitigation. Geo Journal 31 (4), 313-318

Tegegne S D 2012 Livestock water productivity (LWP) improvement in the mixed crop-livestock system of Ethiopian highlands, Amhara region: a gendered sustainable livelihood approach to target LWP interventions for rural poverty reduction. A PhD thesis (unpublished), Rheinischen Friedrich-Wilhelm’s-University of Bonn, Germany.

Thomas C 2011 Drinking water for dairy cattle: part 2: water is the single-most important nutrient for dairy cows. (Accessed on November 28, 2011).

Uddin M M, Sultana M N, Ndambi O A, Hemme T and Peters K J 2010: A farm economic analysis in different dairy production systems in Bangladesh. Livestock Research for Rural Development. Volume 22, Article #122. Retrieved February 20, 2015, from

Uddin M M 2011 Possibilities for further development of market-oriented dairy production systems in Bangladesh. A PhD dissertation. Department of Animal Breeding in the Tropics and Sub-tropics, Humboldt University of Berlin, Germany. Veralag Dr. Köster, Berlin. ISBN 978-3-89574-772-4.

Uddin M M, Sultana M N and Peters K J 2013: Participatory Rural Appraisal to characterize dairy production systems in Bangladesh.Livestock Research for Rural Development. Volume 25, Article #29. Retrieved March 15, 2015, from

Uddin M M, Sultana M N, Brümmer B, Peters K J 2012 Assessing the Impact of Dairy Policies on Farm-Level Profits in Dairy Farms in Bangladesh: Benchmarking for Rural Livelihoods Improvement Policy. Journal of Reviews on Global Economics 1: 124-138.

Van Breugel P, Herrero M, Van De Steeg J and Peden D 2010 Livestock water use and productivity in the Nile Basin. Ecosystems 13:205-221

Zaedi M S, Demura K, Yamamaoto Y, Masuda K Sawauchi D and Nakatani 2009 Bangladeshi Dairy Farmers' Conditions under Milk Vita. The Review of Agricultural Economics, 64, 181-190

Received 23 August 2015; Accepted 2 December 2015; Published 2 January 2016

Go to top