Evaluation of determinants affecting technical efficiency of dairy farms

Abstract

This paper highlights the determinants that influence the technical efficiency of dairy farms in the Chhattisgarh plains of India. For the present investigation, 44dairy farms were randomly selected and administered with pre-developed questionnaires. The data obtained were analyzed using the Cobb-Douglas production function via FRONTIER 4.1 to identify various determinants and management parameters that affect the technical efficiency of dairy farms. A longer calving conception interval (β = − 0.03), regular faecal examinations (β= − 0.08), and the presence of a loose housing system (β= − 0.06) had significantly negative influences (P < 0.10) on technical efficiency. It was concluded that measures to improve peak milk yield (β = 0.58), lactation persistency (β = 0.82), strategic culling of animals (β = 0.03), regular deworming (β = 0.10), high labor productivity (β = 0.15), and improved ratios of pregnant to non-pregnant animals (β = 0.07), forage to concentrate (β = 0.07), and milch to dry animals (β = 0.08) along with the presence of irrigation facilities (β = 0.07), proper dairy waste disposal (β = 0.05) and utilization (β = 0.04), and environmentally controlled dairy housing (β = 0.36), would enhance technical efficiency. The study recommends better feeding, housing, and waste management practices to improve the efficiency of milk production in dairy farms across the Chhattisgarh plains and similar agro-climatic zones of India.

Introduction

To meet consumer demands, there has been tremendous pressure on the livestock sector in general, and the dairy sector in particular, to improve productivity. To address this challenge effectively, improving livestock and their management is one of the viable methods for enhancing productivity. In the Indian context, most dairy owners are small-scale farmers who raise animals using traditional methods¹, which prevents them from fully utilizing available technologies and resources, thereby leading to allocative inefficiencies^2,3. In such cases, improving production efficiency is both feasible and cost-effective, as it allows for maximum output from the given inputs. With this in mind, the present investigation was designed to identify various determinants of dairy farm management, along with their relative contributions, so that appropriate attention can be given to these factors to achieve measurable gains in productivity and profitability.

Materials and methods

Study area, sampling technique & collection of data

The present investigation was conducted in various districts of the Chhattisgarh plain (longitudes 80.5° to 82.5° East and latitudes 21.0° to 21.5° North), which share similar social and environmental conditions. The minimum sample size was determined using Raosoft software⁴. Based on a pilot survey, only 105 dairy farms were included in the sample population, as only these farms were found capable of answering the questionnaire. The confidence level and margin of error were set at 95% and 5%, respectively. The response distribution was just 5%, as most farms were rejected due to lack of interest or incomplete responses. Therefore, a minimum of 44 dairy farms was needed for sampling, so as to generalize the findings. Data from dairy farms located in urban, peri-urban, and rural areas were collected through a simple random sampling approach, using a pre-formed and well-structured questionnaire administered via personal interview method.

Determinants of efficiency

The data (available at Figshare: https://figshare.com/authors/Ashutosh_Dubey/21490085) on various parameters of dairy farm management (production, reproduction, feeding, health, housing, and miscellaneous aspects) were collected from the selected farms. To eliminate the effects of breed and species on herd average, wet average, and peak milk yield, the number of animals was converted into standard animal units (SAU), following the method suggested by⁵. Stochastic frontier analysis (SFA)^6,7 was conducted to determine the levels of production efficiency and inefficiency, using the Cobb-Douglas production function through the FRONTIER 4.1 computer program⁸. To determine the relative contributions of these factors, the technical efficiency scores obtained for milk production were regressed using a simple multivariate regression analysis, as per the following equation:-.

$${\text{T}}{{\text{E}}_{\text{i}}}={\text{ }}{{\text{b}}_0}+{\text{ }}{{\text{b}}_{\text{1}}}{{\text{X}}_{\text{1}}}+{\text{ }}{{\text{b}}_{\text{2}}}{{\text{X}}_{\text{2}}}+{\text{ }}{{\text{b}}_{\text{3}}}{{\text{X}}_{\text{3}}}+{\text{ }}{{\text{b}}_{\text{4}}}{{\text{X}}_{\text{4}}}+ \cdots {{\text{b}}_{{\text{24}}}}{{\text{X}}_{{\text{24}}}}+{\text{ }}{{\text{e}}_{\text{i}}}$$

Where,

TE_i = Technical efficiency of ith farm b₀ = Intercept term.

b₁ to b₂₄ = Coefficient of respective factors affecting production efficiency.

e_i = Random error term.

Table 1 Various qualitative variables considered for study.

The model fit (R²) for the regression equation was evaluated using SPSS software. Among the 24 variables included in the analysis, several exhibited multicollinearity—specifically, peak milk yield (variance inflation factor = 31.87), herd average (VIF = 60.23), labour productivity(VIF = 11.43), regularity in faecal examination (VIF = 10.40), presence of irrigation facilities (VIF = 14.63), forage concentrate ratio (VIF = 15.00), automation in milking (VIF = 17.61), and degree of farm automation (VIF = 13.31). Despite the presence of multicollinearity, the model demonstrated a high coefficient of determination (R² = 0.99), indicating that 99% of the variation in milk production was explained by the predictors included in the model. This strong model fit highlights the substantial contribution of variables such as peak milk yield, reproductive performance, and various farm management indicators. It should be noted, however, that a comparison of basic herd descriptors between responders and non-responders was not performed, due to data unavailability for non-responding farms. This is acknowledged as a limitation and may introduce potential non-response bias in the study.

Hypothesized scores

Table 1 depicts the various qualitative variables considered in the study, which cannot be evaluated in terms of absolute values. For such parameters, deviations from the ideal management values were scored using dummy variables, following the methods described by^9,10,11,12.

To balance the reproduction and production cycles, it was assumed that 65% of the breedable population should remain pregnant in a herd, along with an adequate proportion of milch animals (70% for indigenous breeds, 75% for crossbreeds, and 65% for buffaloes). Similarly, a dairy farm with cattle-to-buffalo ratio of 40:60 was considered more profitable than maintaining a single species herd^13,14.

Automation in milking is believed to improve milk production and reduce labor requirements by 12% and 18%, respectively¹⁵. To ensure a consistent supply of green fodder throughout the year, dairy farms should maintain adequate stocking capacity. The sound herd health management practices viz. regular deworming and vaccination, helps to prevent unnecessary expenditures on treatment and reduce both morbidity as well as mortality rates among farm animals.

The concentrates are expensive but are rich sources of protein and energy. However, to align with ruminal physiology, roughages are essential to enhance dry matter intake and milk production. Therefore, an adequate combination of roughages and concentrates is necessary to meet the nutritional needs of dairy animals. On a fresh weight basis, 75% of roughage should be provided as green fodder and 25% as dry fodder^16,17,18.

The behavior and welfare of dairy cows are influenced by the physical environment in which they are housed¹⁹. Loose housing systems allow animals to express natural behaviors and are therefore important for animal welfare²⁰. Furthermore, environmentally controlled dairy housing helps protect animals from extreme environmental conditions. The presence of a waste disposal system and utilization practices not only reduces environmental and health hazards but also generates additional income for the farm.

Results and discussion

The raw data, along with their log-transformed values, obtained from this study, are provided in a supplementary table. Moreover, the statistical analysis viz. Instructional file, input and output file for determination of technical efficiency, was submitted as related files.

Table 2 Descriptive statistics of variables under consideration.

Table 3 Technical efficiency of farms on the basis of standard animal units (SAU).

Table 2 presents the summary/descriptive statistics of the variables used in the study. Since, the data was collected from rural, urban and peri-urban areas for maintaining the randomization of data, the same degree of variability was observed in different variables and among various management profiles of the farms.

The technical efficiency of farms with small standard animal unit (SAU) sizes was relatively high as compared to those in the medium and large categories. In contrast³, reported higher efficiency for large farms (89.71%) than for small and medium dairy farms. To achieve maximum efficiency, all categories of farmers would need to realize, on average, a cost saving of 4.21% (Table 3). The high mean technical efficiency observed across various dairy farms may be attributed to the inclusion of a large number of independent variables, which could allow the inefficiency of one variable to be masked by the efficiency of another within the same farm. ²¹had also observed high technical efficiency for dairy farms (Mean 95%) in Gaziantep region of Turkiye.

The small farms may be more efficient because they have simpler management needs, fewer animals to coordinate, and better resource allocation (e.g., labor intensity and feed access). Although small dairy farms face resource constraints and various economic limitations, the smaller herd size allows for better human-animal interaction, which leads to greater attention to animal welfare. ^2,22also observed high technical efficiency of 94.57%, 84.40%, and 92.62% for small, medium, and large herd size categories, respectively, while working in the plain region of Uttarakhand, India, supporting the findings of the present study. Furthermore²³, using the Data Envelopment Analysis (DEA) approach, reported a negative association between herd size and production efficiency. However²⁴, using DEA and²⁵ using SFA had independently observed lower efficiency among smallholders and a non-linear relationship between herd size and efficiency, respectively.

Table 4 The final maximum likelihood estimates (MLE) of various parameters.

In Table 4, the estimated variance ratio (γ) was found to be 0.99, which is very close to 1, hence statistically significant at the 1% level. This indicates that variations in milk production across different dairy farms were primarily due to differences in technical efficiency. The remaining 1% of variance in milk production was attributed to random noise or error, these are the factors beyond the control of the farmers. The Sigma², an error term (u), another measure of technical efficiency, was a predominant error. These errors were farm-specific and directly under the control of the farmers, and could be minimized through various farm-level management interventions. A high level of technical inefficiency in milk production was also observed by^26,27,28.

The factors found to have a significant influence on the technical efficiency of milk production include calving conception interval (CCI), first service conception rate (FSCR), peak milk yield (PMY), lactation persistency (LP), culling/mortality rate (CR), labor productivity, pregnant non-pregnant ratio (P: NP), milch dry ratio (M: D), regularity in deworming (RD) and faecal examination (RF), forage concentrate ratio (F: C), presence of waste disposal and utilization systems (WDS & WDU), irrigation facilities (IF) at the farm, and housing systems that are loose and environmentally controlled (ECDH) (Table 4). High coefficient values were observed for peak milk yield (0.58) and lactation persistency (0.82), meaning that an increase of one unit in either of these variables may increase milk production by 0.58 and 0.82 units, respectively. This may be because both variables directly contribute to increased total milk yield on the farm.

The high milking status tends to pose animals into a state of negative energy balance (NEB). This postpartum negative energy balance affects the follicular fluid composition, thereby delaying the formation and maturation of oocyte along with delaying the recovery of ovarian activity, prolongation of first follicular wave and reduces the rate of ovulation²⁹. These physiological responses may explain the negative association between first service conception rate and calving conception interval with technical efficiency of milk production. In contrast, increased peak milk yield also enhances the 300-day total and overall lactation yields^30,31], thereby improving both profitability and technical efficiency of dairy farms¹⁶.

The persistency was positively correlated (0.18 and 0.057, respectively) with decrease in the rate of milk yield^32,33 and with a larger calving conception interval³⁴. Farms with high technical efficiency in milk production tended to have longer calving conception intervals, better health status, optimal body condition scores, and lower disease incidence³⁵ all of which may contribute to increased persistency.

The high productivity may negatively impact the health status and fertility of the animal^36,37. Moreover, in specialized dairy farms, it is a common practice to improve profitability by reducing costs associated with sick or non-pregnant animals³⁸.

The labour cost is second most significant operating expense in running any dairy farm³⁹. Dairy farms that utilize labour efficiently i.e. with lower investment in labor costs and higher milk output, tends to achieve greater labor productivity and farm profitability⁴⁰. This finding is consistent with⁴¹, who observed a significant association between milk productivity and labour use efficiency. Similar association of labour wages on efficiency was also observed^42,43,44 using Stochastic frontier production analysis and by⁴⁵ using Data envelopment analysis.

The optimum ratio of pregnant to non-pregnant animals indicates a proper balance in the number of milch cows at different stages of lactation. Failure to maintain this ratio increases the likelihood of involuntary culling, raises heifer replacement costs, and reduces the number of calves born per cow⁴⁶.

The drying off-calving-initiation of a new lactation constitutes a series of transitional events that occur in a well managed dairy farm. Such a sequential event is essential for maintaining sustainable production, profitability, and efficiency. The present finding is consistent with^47,48, who observed a significantly positive effect of the number of lactating animals on the technical efficiency of dairy farms.

The regularity in deworming not only enhances the feed utilization capacity of farm animals but also improves their health status, thereby, improving both the quantity and quality of milk. This finding is consistent with⁴⁹. However, regular faecal examination requires additional labor and financial investment due to underlying health concerns. Moreover, it may divert the manager’s attention from other important efficiency-enhancing managerial activities on the farm. As a result, it has a significantly negative impact on the technical efficiency of milk production. ²¹ also observed significant effects of health management practices on the technical efficiency of milk production.

The presence of irrigation facilities on a dairy farm ensures a continuous supply of water for fodder production, in addition to its use for drinking, cleaning animals, and maintaining hygiene in housing areas. Thus, adequate fodder production per adult livestock unit (ALU), along with proper cleaning of the dairy house, was reflected in this association. Similarly, an appropriate forage concentrate ratio ensures the balanced supply of essential nutrients necessary for maintaining the health, productivity, and reproductive efficiency of the animals. Animals receiving adequate dietary fiber were found to have better feeding rates, higher dry matter intake, and improved milk yield, along with reduced rumination time and enhanced lipid metabolism^50,51,52. also reported a significantly positive effect of feeding practices i.e. including hay, silage, concentrate, and hay ratios, on the technical efficiency of milk production. Furthermore⁵³, had also observed the significant effects of green fodder and concentrate on efficiency of milk production.

The dairy waste materials, particularly, dung and urine, pose a biological risk due to their high bacterial load. They also raise welfare and environmental concerns by contributing to greenhouse gas emissions such as methane, nitrous oxide, and ammonia, which adversely affect animal health and productivity^54,55,56. These concerns are significantly reduced in farms with effective waste disposal systems. Moreover, utilizing dairy waste for crop production not only generates additional income for farmers but also contributes to environmental sustainability, which was reflected in the form of positive association with milk productivity and technical efficiency of milk production. The potential for extra revenue generation from dairy waste was also reported by⁵⁷.

Despite the various benefits of the loose housing system, as hypothesized in the Materials and Methods section, this system is associated with a higher incidence of lameness⁵⁸, greater exposure to unpredictable weather conditions, and an increased risk of infections and endoparasitic infestations. Additionally, animals reared in pasture-based housing systems experience more severe negative energy balances and various metabolic disturbances during the postpartum period. These metabolic and nutritional stresses negatively affect the animal’s immune function, production potential, and fertility.^59,60 also reported a longer calving conception interval among cows reared in increased space compared to those kept in confined housing. In contrast to this,²³ had observed the significantly positive effects of scientific animal housing on efficiency of production, using DEA approach.

The behavior and welfare of dairy cows are influenced by the physical environment in which they are housed¹⁹. Environmentally controlled dairy housing supports not only sustainable milk production but also the longevity of dairy cows, which in turn reflects a balance among production, reproduction, and overall health. Therefore, it contributes not only to animal welfare but also to the quality and profitability of sustainable dairy farming.⁶¹. ²³also reported significant effects of scientifically designed animal sheds on the technical efficiency of milk production.

The findings suggest that it is important to balance the production and reproduction status of dairy animals by maintaining proper feeding and housing standards to address their welfare and health needs. Measures to improve peak milk yield, lactation persistency, strategic culling, regular deworming, labor productivity, and balanced ratios of pregnant to non-pregnant animals, forage to concentrate, and milch to dry animals, along with the presence of irrigation facilities, waste disposal and utilization systems, and environmentally controlled dairy housing—may increase technical efficiency. Training and workshops should be conducted for dairy farm managers to support better utilization of all available resources, including human resources and dairy-generated waste materials.