Estimation of Technical Efficiency in Tea Farms at Plantation Level: A Review

The systematic literature review of 111 abstracts has been conducted to comprehensively compile the empirical studies of 21 complete text papers from all over the globe in context to estimation or determination of the technical efficiency (TE) at plantation level of tea production system (TPS), by adopting two methodologies viz ., stochastic frontier analysis (SFA) and data envelopment analysis (DEA) during the period 2012-2022. Investigation from these empirical studies revealed that the average TE (TE mean ) tea growers TGs all around the globe computed by using both the approaches is around 67.98%, which showcased that the TGs have ability to increase the green tea leaf (GTL) production by 32.02% through better utilization of available resources and technology. The influence of various factors on TE of these TGs had contradictory outcomes, which broke new ground for future research. Computation of TE will enable an investigator to benchmark the best performing TGs in a particular area, which may be referred by the inefficient TGs to enhance their performance.


Introduction Production Efficiency as a Powerful Tool in Measuring the Performance of a Tea Garden
Like any other agricultural production, the tea production at plantation level involves transformation of some goods and services called input into other goods called products or output. 1 Agricultural productivity (AP) is defined in agricultural geography as well as in economics as "output per unit of input " or "output per unit of land area", and the improvement in AP is generally considered to be the results of a more efficient use of the factors of production, viz. physical, socioeconomic, institutional and technological. 2 The AP depends on two components, which are as follows. 3

Production Technology (PT)
It is characterized by the type and quality of inputs and resources used in the production process. For a given commodity like tea, many different technologies may exist, reflecting different economic, environmental and agronomic conditions.

Technical Efficiency (TE)
It refers to ability of the production process to combine the available resources or inputs to produce maximal output (GTL). A tea farm is technically inefficient when it does not produce the maximum level of output that can be expected given the type of available inputs.
The productivity variation has a significant impact on the production of tea in the case of TGs. Productivity in the case of any STG is defined as the "yield of tea grown per hectare or per area of land". 1 The overall efficiency (OE), consisting of both the consisting of both the TE and allocative efficiency (AE) 1 of individual TG 4 is known as the economic efficiency (EE) of the individual TG. It refers to the ability of the TG to minimize the cost of cultivation without altering the desired yield of GTL from the farms. Lowering costs while preserving productivity means higher profits, which is why EE is a common strategic goal. 5 In practice, a technically efficient farm can be economically inefficient, whereas the reverse may not be true. It is especially true in developing countries where markets are often thin or inexistent, inputs are constrained (unavailable or difficult to access), and transaction costs are high. 3 The term "efficiency (η)" signifies a peak level of performance that uses the least amount of inputs to achieve the highest amount of output. Efficiency analysis serves as one of the most powerful tools to understand how inputs are translated into valued outputs. 6 An efficient TG will reduce the number of unnecessary resources used to produce a given output (GTL), including personal time and energy. Efficiency is a measurable concept that can be determined using the ratio of useful output to total input. It minimizes the waste of resources such as physical materials, energy, and time while accomplishing the desired output. 7

Benchmarking -A Technique for Establishing Gaps in Performance of a Tea Garden
The word 'benchmark' originated from a surveyor's mark cut to indicate a level for the determination of altitude. 8 Benchmarking of tea productivity (TP) may be considered as a management technique, in which measurement is primarily comparative. A TG could attempt benchmarking at several levels using all the different types of benchmarking with the purpose to find out the best practices so that it could confirm to it. Typically to benchmark TP among a homogenous set of TGs, the "best practice benchmarking or process benchmarking" technique is generally applied to compare the methods and practices for performing tea production processes. 9 Our study mainly focuses on the benchmarking TP of TGs on the basis of TE, assuming that the TGs use same quality of inputs and resources in the production process. The optimal productivity target which has to be compared to observe TP to measure the degree of TE {or technical inefficiency (TI)} at the farm-level is theoretically known as the production frontier. 3 From the definition of TE, it is clear that it is a relative measure, not an absolute measure and can be measured by two different ways viz. output oriented technical efficiency (TE oo ) and input oriented technical efficiency (TE io ). 10

Methods for Benchmarking of Tea Farms
In modern benchmarking the two main approaches are SFA and DEA. 11

Parametric Stochastic Frontier Analysis (SFA)
The first prominent concept on modeling and estimation of SFA is forwarded by the empirical work of the concept of a stochastic production frontier (SPF) was developed and extended by Aigner, Lovell, and Schmidt in 1977. Further, Battese and Coelli in 1995, Greene in 1990 and Wim and Broeck in 1977 provided a significant contribution for the progress of SFA considering different distributional assumption of the error term. 12 Typically, the production or cost model is based on a Cobb -Douglas (CD) function 13 or translog (TL) function. 4 Based on the different distributional assumption of the error terms, the SFA approaches can be modeled in the different ways viz., Half Normal model, Truncated Normal model, Exponential model 14 and Gamma model. 15 The production function under stochastic frontier distinguishes the error term associated with the production function in to statistical noise and inefficiency components. It is assumed that each component has their influence in deviating output from the most possible maximum level. The statistical noise or uncontrolled component is the error due to randomness which is two-sided. For example noise components like weather, climatic condition or any unexpected event may either increase or decrease the yield of tea in the tea farm, which is beyond the control of the cultivator. On the other hand, production inefficiency component is only due to inefficiency in allocating resources which is a one-sided error. This error has a negative impact on the production function and can be controlled by the cultivator with appropriate measures. 12 The focus of all 15 papers (71.43%) utilizing the SFA techniques is to examine the efficiency levels of TGs through the estimation of TE. 16 has foreground benchmarking of TGs of Vietnam on the basis of Resource Use Efficiency (RUE) of these by application of this method. In addition to the estimation of TE, 17 used the SFA cost function to find out the reason for variation in the Total Cost of Production (TCP) of GTL, Further the researcher calculated the EE of two different sets of small tea growers (STG)s and used the Mann-Whitney U Test (Wilcoxon Rank-Sum Test) to find the difference in the efficiency levels of the two independent groups of STGs. Similarly, 18 used the Aigner et al. (1977) and Meeusen & Broek (1977) SFA production and cost functions in CD form to calculate the TE mean and average Cost Efficiency (CE) 2 {CE mean } scores of the levels of organic STGs respectively. The researchers subsequently used these values to estimate the EE levels of the set of STGs.

Non -parametric Data Envelopment Analysis (DEA)
Initially put forward by Charnes, Cooper and Rhodes in 1978 and further enhanced by Banker, Charnes and Cooper in 1984, 19 DEA being a nonparametric approach does not require a functional form specification and is easy to compute using linear programming. In case of tea industry, it considers each tea farms (termed as 'decision making units' or DMU) and calculates a discrete piecewise frontier determined by the set of efficient tea farms or best practice units. It makes a comparative analysis of the tea farms that utilizes multiple inputs to produce multiple outputs which can be quantified using different units of measurements. Each DMU has the flexibility with respect to some of the decisions it makes, but not necessarily complete freedom is given with respect to these decisions. This method cannot separate the effect of noise and effects of inefficiency during the calculation of TE, and is less sensitive to the type of specification error. The most popular models of DEA widely used to carry out research work 20 are as follows

CCR Model
This was the first DEA model was suggested by Charnes, Cooper and Rhodes in the year 1978 and is based on the constant return to scale (CRS) assumption. The efficiency measured under CRS assumptions represents the technical efficiency (TE or TE CRS ).

BCC Model
Banker, Charnes and Cooper (1984) further extended the work of Charnes, Cooper and Rhodes, keeping into consideration the various factors might cause a tea farm to deviate from its optimal scale of operations, thus accounting for variable returns to scale (VRS). The efficiency measured under CRS assumptions represents the pure technical efficiency (PTE or TE VRS ).
Both the models are used simultaneously in various empirical studies on TE of tea farms to estimate their scale efficiency (SE) 3 . The focus of the majority of the studies by application of DEA is on evaluating TE scores. However, 21 and 22 used DEA method for determination of SE scores in addition to TE scores. Later on 23 benchmarked TGs of Turkey on the basis of TE, SE, AE, EE and PTE scores by application of this method.

Determinants of Technical Efficiency
The various factors affecting the TE of TGs can be determined using different models of Multiple Regression (MR). 4 Our studies revealed that the TE of the TPS at plantation level is dependent on numerous factors which are stated as follows.

Objective
The intention of this study is to find the limitations and ambiguity in the existing investigations carried out in estimation or determination of TE of TGs, and subsequently finds the impact of various factors on the TE of these TGs. These significant research works may be referred by other researchers to investigate the performance of the STGs in other unexplored regions of the world, where there has been no study conducted so far. Also, a significant statement was made by the erstwhile Commerce Secretary, Government of India that the tea industry should benchmark itself against best practices so that it can compete in international market against countries like Kenya and Sri Lanka. 24 This will help the industry to solve its fundamental challenges on decline in productivity. With a steep hike in the input cost, the tea industry should make an attempt to utilize the available resources judiciously, i.e., without making any wastage of the resources and achieve the optimal level of production for its self sustainability in the competitive environment.
It may be noted that these literatures will enable the researchers to identify the factors which are responsible for causing the (in) efficiency in the tea production system and subsequently adopt strategies to rectify the same.

Methodology
To systematically highlight the quantity, status of research work done, and the scope for the future research, an investigation for the TE for the tea sector was searched from all the accessible/ available published paper using "technical efficiency (of) OR (in) tea" as the phrase with the above mentioned keywords in the academic search engine Google Scholar was used for retrieving relevant literature. In addition to this records were identified through other sources. From retrieved literature, relevant investigations carried out during the period 2012-2022 in top tea producing counties in the world were taken into consideration. The studies which investigated the TE of only the TGs {special focus on small tea grower/ gardens (STG)} were included for the systematic review, excluding the tea processors and tea estates. The results from the academic search engine and other sources were filtered by using inbuilt advanced searched operators 25 and Boolean operators. 26 The empirical works carried out using only the two common methodologies viz., Parametric SFA and Non -parametric DEA were taken into consideration. The following diagram is the pictorial representation of the systematic literature review methodology using PRISMA flow diagram.      38 35 Interestingly, the study conducted by 35 revealed that it was revealed that there exists an inverted U-shaped (non-linear) relationship between the TE and P AGE , and the turning point of age was found to be 42.813 years. In a recent study conducted in Sri Lanka by 37 found that the TE mean of the organic STGs is 0.247, which is lowest among all the studies conducted in the country. Contrary to this, an investigation carried out by 18 found that the TE mean of the STGs is 0.85, which is comparatively high than the TE mean of the STGs calculated from other studies in the country. A similar study conducted by 33 in the Vi Xuyen district, Ha Giang province of Vietnam revealed that TE mean for conventional tea production (CTP) cultivators {(TE mean ) CTP } (0.701) higher than that of and organic tea production (OTP) cultivators {(TE mean ) OTP } (0.652). it is noteworthy that, 33 adapted the Discrete Choice models in form of Binary Logit to determine the influencing factors of the tea farmer's choice (decision) on OTP.
The ambiguous outcomes related to the impact of organic conversion of tea farms on its TE in different nations will thus create a dilemma on the farmer's decision to adopt organic tea farming.

Interpretation of the Studies Using Dea Technique
Out of 21 studies reviewed it was found that 5 (23.81%) studies used DEA to determine the TE of the TGs at plantation level as stated in Table (2) above. The plantation level TE mean of TGs determined by this method is 0.676617. The studies reflected the use of both BCC and CCR models of DEA. The use of different models of DEA by researchers in carrying out their studies is stated below. The various statistical techniques applied by the researchers in their studies to analyze the effect of various factors on the TE of TGs are stated below.

Sl. No. DEA Model Studies
It was notable that 23 conducted the VIF Diagnostic Test to check the multi-co linearity among the independent variables prior to the determination of the factors contributing to the efficiency of the TG farms.
The significant contrasting outcomes from the studies using DEA technique is stated below, which puts forth further avenues for research.

Interpretation of the Studies using both SFA and DEA Technique
The only study carried out by 42 in India reflected that the TE mean of the STGs determined by using DEA technique {(TE mean ) DEA } is 0.8167 which was higher than that by using SFA {(TE mean ) SFA }in the same set of data (0.62), as it takes into account the data noise such as errors and omitted variables. The study also found that the means of procurement of leaf from STGs by the large tea estates (BTG) has been adopted on the ground of CE or cost of production (COP) of GTL.

Conclusion
It can be concluded that to raise the efficiency and productivity of the tea farms, it becomes imperative to quantitatively measure the existing level of TE and policy options available for raising the present level of efficiency, given the fact that efficiency of production is directly related to the overall productivity of the plantation sector. The empirical evidence is very important in identifying the factors that threaten the productivity of these units and in generating information for designing of support policies for the small tea gardens and institutional improvement.

Future Outlook
• The impact of various factors on their TE had contradictory outcomes in different studies.
To validate such contradictions, a further investigation is required to be carried out to find the impact of such factors on the TE of the tea farms in different geographical locations.

•
The quality parameters of the tea farms were not taken into consideration in any of the studies to benchmark the best practicing farm, which opened up the scope to carry out investigation incorporating the quality aspect of the tea farms to measure the performance of the farms.