Literature review

There is abundant literature discussing different aspects of conventional, organic, or inorganic farming or comparisons of it. Gimeno-García et al. (Reference Gimeno-García, Andreu and Boludab1996) studied the use of other inorganic fertilizers and pesticides to impact soil fertility in Valencia, Spain. Their study was based on field experiments. It concluded that although a different type of inorganic fertilizers and pesticides does influence soil, their influence is lower than the standard set by the European and Spanish legislation.

The comparison between different types of farming is extensively undertaken in scholarly literature. However, much of this work discusses organic and conventional farming while considering various aspects. For instance, Gabriel et al. (Reference Gabriel, Sait, Kunin and Benton2013) studied the cost (measured in terms of reduction in agricultural yield) and benefits (measured in terms of biodiversity) for agrarian management in their study of the relationship between organic and conventional farming in England. They found a reduction of biodiversity with increasing crop yield and about 54% lower production gain for organic fields than the traditional field. Similarly, Gomiero et al. (Reference Gomiero, Pimentel and Paoletti2011) also compared organic and conventional farming and considered several different aspects to both types of farming, such as soil characteristics, yields, CO₂ abatement, impact on biodiversity and energy uses. They concluded that organic farming is a better option than conventional farming. Mondelaers et al. (Reference Mondelaers, Aertsens and Van Huylenbroeck2009) also reported that organic farming is positively linked with agro and biodiversity with increased production per unit of land compared to conventional farming.

Biodiversity and types of farming is also a focused area of scholarly literature. For example, Maeder et al. (Reference Maeder2002) reported findings of 21 years long field studies that were set up to study biodiversity and soil fertility of organic and conventional farming. They said lower yield with enhanced soil fertility, increased biodiversity and less dependent on external inputs such as fertilizers and energy for organic farms. Similarly, Gabriel et al. (Reference Gabriel2010) studied the impacts of farms types and land use at multi spatial scales (field, farm and landscape) on biodiversity. They reported that at the farm level, biodiversity depends on the farming choice between organic or conventional. A review study by Hole et al. (Reference Hole2005) focused on the impact of organic and inorganic farming on biodiversity. They compared various studies relative to biodiversity and organic farming. Hole et al. (Reference Hole2005) reported a wide range of taxa (such as birds, animals (mammals and invertebrates) and arable flora) that can be benefited from organic management through an increase in abundance and species richness. Furthermore, the study also emphasized reduced use of chemical pesticides, inorganic fertilizers and sympathetic control of non-cropped habitats.

Some other aspects, such as environmental factors, land disturbance and water usage, also focus on scholarly literature while comparing various types of farming. For instance, Wood et al. (Reference Wood, Lenzen, Dey and Lundie2006) compared conventional and organic agriculture in Australia, using some factors such as energy requirement, land disturbance, water use, employment and greenhouse emission. Their results suggested that organic farming had higher energy use and greenhouse gases emission compared to conventional agriculture. However, organic farming is better for food production than traditional farming while considering direct water usage, labor input and local benefits. Norton et al. (Reference Norton2009) compared the consequences of non-organic and organic farming on habitat and management for a total of 250 farms land from England. Their study concluded that organic farms were more heterogeneous in the landscape, with greater field and farm complexity than non-organic farms.

Studies that compare the choice between organic and inorganic farming is scarce. However, some studies on organic and inorganic farms comparison for grapes (Waykar et al., Reference Waykar, Yadav, Shendage and Sale2006), rice and wheat (Sujatha et al., Reference Sujatha, Eswara Prasad and Suhasini2006) in India. Waykar et al. (Reference Waykar, Yadav, Shendage and Sale2006) reported a benefit-cost ratio of 1.37 and 1.52 respectively for grape organic and inorganic farms in Maharashtra, India, using a sample of 60 farms. Sujatha et al. (Reference Sujatha, Eswara Prasad and Suhasini2006) did a similar kind of comparison using a multi-stage random sampling technique based on primary data collected from rice and cotton farms in Andra Pradesh, India. They reported higher labor costs with a higher return for both crops in organic farming.

In contrast, the cost of fertilizers and pesticides was higher for inorganic agriculture, with a lower return for both crops than organic farming. Epule (Reference Epule, Chandran, Unni and Thomas2019) is a review study that, based on already published literature, presents the conceptual issues related to organic and inorganic farming followed by both types of farming role on global food security. They concluded that organic agriculture could reduce global food insecurity, but beyond a certain point, it's the combination of both types of farming that can result in a better outcome.

Norton et al. (Reference Norton2009) is based on a field survey from farmers obtained from 250 farms lands and a comparison of that information. Gabriel et al. (Reference Gabriel, Sait, Kunin and Benton2013) used the landscape of equal sizes for organic and conventional farms and then clustered them for comparison purposes. A similar kind of methodologies was also adopted by Gabriel et al. (Reference Gabriel2010) using a multi-scale hierarchical sampling design. Gomiero et al. (Reference Gomiero, Pimentel and Paoletti2011) employed a descriptive approach in which different factors such as soil characteristics, yield and CO₂ abatement were compared for organic and conventional farming using the evidence available from published scientific and scholarly studies. Maeder et al. (Reference Maeder2002) is based on 21 years of field studies that compared biodiversity and soil fertility from conventional and organic farming. Wood et al. (Reference Wood, Lenzen, Dey and Lundie2006) used input-output based life cycle analysis to compare traditional and organic farming.

Some studies do discuss organic, conventional and inorganic farming in Pakistan. For example, Waqas et al. (Reference Waqas, Hong, Maruf, Rana, Saima and Ali2017) reported that organic and conventional farming are the two most common types of farming in Pakistan. Farming is mainly dependent on the use of chemical fertilizers and pesticides. It is also important to highlight that not all animal can digest phytates that are part of livestock feed and as a consequence not digested phytates ends up in manure and later in soil (Sun et al., Reference Sun2017). This manure, if used as a fertilizer, can have harmful ecological impacts (Wu et al., Reference Wu2015). However, farmers are hesitant to switch to organic farming because of expect a lesser return and are not fully aware of the benefits of organic farming.

Furthermore, some pesticides such as Glyphosate is used to kill unwanted plants (herbicides) but the use of glyphosate for a longer period could have severe health outcomes for humans (e.g., Jaisi et al., Reference Jaisi2016; Li et al., Reference Li2018; Sun et al., Reference Sun, Li and Jaisi2019). Also, the heavy usage of pesticides and chemical fertilizers reduces soil fertility, organic matter and crop production (Waqas et al., Reference Waqas, Hong, Maruf, Rana, Saima and Ali2017). Furthermore, both approaches vary from case to case within rural-urban areas. In rural areas, farming is based on a more conventional method than in urban areas. However, Waqas et al. (Reference Waqas, Hong, Maruf, Rana, Saima and Ali2017) is based on a field survey and use only mean statistics.

It may be noted that the findings of scholarly literature (from Pakistan) report that inorganic farming is more profitable for farmers compared to organic farming. For instance, ul Hussain and Khan (Reference ul Hussain and Khan2017) did a field survey of 444 farms (rice and wheat) consisting of 220 organic and 224 inorganic farms in three districts of Punjab province of Pakistan. Their survey reveals that farmers had concerns about the market for organic products being not developed, organic inputs not being readily available, lesser support from the government and lack of zoning. They applied statistical tests while using factors like soil's health, cost of input, profitability and yield from both methods. They concluded that conventional and organic farming are equally profitable, with organic farms showing lesser input costs.

Additionally, organic farming converse soil fertility. Mehmood et al. (Reference Mehmood, Anjum, Sabir and Arshad2011) used a survey from randomly selected villages within the district of Sheikhupura, Punjab. They applied descriptive analysis and statistical tests and concluded that net income from inorganic farming was higher than net income from organic farming. The study links these findings to the lower cost for inorganic agriculture, unavailability of the market for organic products and smaller landholding by farmers. The benefit-cost ratio for organic farming was 1:1.08, and for inorganic, it was 1:1.01.

The studies explored in this section used different aspects of traditional, organic and inorganic farming. It is clear from the literature survey that there is little scholarly work that compares organic and inorganic farming, even though there are many comparison studies between conventional and organic farming. Also, the factor considered for comparisons are mainly biological (e.g., Maeder et al., Reference Maeder2002; Gomiero et al., Reference Gomiero, Pimentel and Paoletti2011) or are focused on biodiversity and soil fertility. Hardly any of these studies combine biological, social and economic factors in a single research and consider farmers and consumers. Farmers are the essential stakeholder in developing countries; however, there is hardly any study that discusses whether replacing conventional farming with organic farming is economically viable for the farmers. If organic farming can provide sound environmental and health effects (and farmers know that too), but if it cannot give the farmers their required economic return, it may be not attractive for the farmers to adopt organic farming. This aspect of organic and inorganic framing is not explored fully. Finally, despite the usage of methods comparison (ranging from field surveys to the meta-analysis and input-output based life cycle analysis) being applied to compare various types of farming, none of the earlier studies uses multi-criteria decision-making techniques.

The current study contributes to the literature some of these identified gaps by providing a comparison of organic and inorganic farming. The study combines social, economic aspects of farming. The study also considers farmers and consumers, and factors that may influence the choice between organic and inorganic farming. The study also uses modern decision-making techniques based on multi-criteria decisions making as explained and undertaken in subsequent sections.

Methodology

Analytic hierarchy process (AHP)

AHP was developed by Saaty (Reference Saaty1980). This method has become popular in recent years for ranking and selecting the best alternative considering various criteria. More recent literature on AHP is refugee studies (Ali et al., Reference Ali, Sabir and Muhammad2019), location choice (Ali et al., Reference Ali, Butt, Sabir, Mumtaz and Salman2018) and energy optimization (Ali et al., Reference Ali, Butt, Sabir, Mumtaz and Salman2018). According to (Ali et al., Reference Ali, Sabir and Muhammad2019), AHP can be applied in the following four steps;

Step 1

In this step, the comparison matrix [C] is completed using rating as equally preferred to 1, moderately preferred to 3, strongly preferred to 5, strongly preferred to 7 and significantly preferred to 9. A value of 1 is assigned if the alternative is compared with itself. If a choice is preferred over the second alternative, then the weight of the second alternative will be the reciprocal of the first alternative.

Step 2

In this step comparison matrix [C] matrix is converted to the normalized matrix [Norm C]. To find the normalized weight (W_i) of each criterion by using the normalization of the geometric mean of rows in the comparison matrix

(1)$${\rm H}{\rm N}_{\rm i} = (\textstyle \prod\nolimits_{{\rm i} = 1} {^{\rm M} } \,{\rm x}_{{\rm ij}})^{1/m}\;{\rm and}\;{\rm W}_{\rm i} = {\rm H}{\rm N}_{\rm i}/\sum\nolimits_{{\rm i} = 1} {^{\rm M} } \,{\rm H}{\rm N}_{\rm i}$$

Step 3

Average row values. This is the Criteria Weights vector {W}

Step 4

Step 4 consist of several small things. First consistency check on [C] is performed. And then, a Weighted sum vector {Ws} is calculated by multiplying the consistency ratio (CR) with the criteria weights vector. Afterwards, the consist vector {Cons} is calculated by dividing the weighted sum vector over the standards weights vector. Now the value of λ as the average of values in the consistency vector is calculated. Finally, the consistency index is evaluated and are used to calculate the CR. It may be noted that if the CR is less than 0.1, then the criteria weight vector is valued. Otherwise, an adjustment in C is required to undertake until the CR becomes consistent [for more details on it, please see Ali et al. (Reference Ali, Sabir and Muhammad2019)].

Technique for order preference by similarity to ideal solution (TOPSIS)

TOPSIS was first developed by Hwang and Yoon (Reference Hwang and Yoon1981). It is one of the best methods of MCDM. Its basic principle is that the chosen alternative should have the shortest distance from the positive ideal solution and the farthest distance from the negative ideal solution. This technique helps to find the best ideal alternative among different conflicting criteria.

The different steps of TOPSIS, as also listed in (Ali et al., Reference Ali, Sabir and Muhammad2019), are listed below.

Step 1

Construct of the decision matrix that is expressed below in Equation (2):

(2)$${\rm D} = \left({\matrix{ {d11} & \cdots & {d1m} \cr \vdots & \ddots & \vdots \cr {dn1} & \cdots & {dnm} \cr } } \right)$$

Step 2

The normalized decision matrix is constructed in this step

(3)$$r_{ij} = \displaystyle{{d_{ij}} \over {\sqrt {[ \mathop \sum \nolimits_{i = 1}^m d_{ij}]\; \^{}\;{2}} }}\quad i = 1 \ldots \ldots \ldots m\quad j = 1 \ldots \ldots \ldots n$$

Step 3

Construct a weighted normalized decision matrix; each column of the normalized decision matrix is multiplied by its associated weight. Assume that the weight of each criterion is wj for j = 1 2……n

The element of the new matrix become:

(4)$$V_{ij} = W_j \times r_{ij}$$

Step 4

Ideal Positive and ideal Negative solution is determined from;

Positive Ideal Solution:

(5)$${\boldsymbol A}^{\boldsymbol \ast } = \{ {v_1^\ast \;, \;\;\ldots , \;\;v_n^\ast } \} , \;$$

Where $V_j^\ast \; = \{ max\;( {v_{ij}} ) if\;j\in J; \;\{ min\;( {v_{ij}} ) \;if\;\;j\in J^{\prime}$

Negative Ideal Solution:

(6)$${\boldsymbol {A}^{\prime}} = \{ {v_1^, \;, \ldots \ldots , \;v_n^,} \} $$

Where $V^, = \{ min\;( {v_{ij}} ) if\;j\in J; \;\;\{ {max\;( {v_{ij}} ) if\;\;j\in {J}^{\prime}} \}$

Step 5

For each alternative, separation measures are calculated;

Positive separation:

(7)$$S_1^\ast = \sqrt {\;\left[\mathop \sum \nolimits_{i = 1}^m {( {v_i^\ast \ndash \;v_{ij}} ) }^2} \right] \quad i = 1, \ldots , \;\;m$$

Negative separation:

(8)$$\;\;S_1^{\prime} = \sqrt {\;\left[\mathop \sum \nolimits_{i = 1}^m {( {v_i^{\prime} \ndash \;v_{ij}} ) }^2} \right] \quad i = 1, \ldots , \;\;m$$

Step 6

Relative closeness to the ideal solution is calculated

(9)$$C_j^\ast = S_i^{\prime} /( {S_j^\ast{ + } S_i^{\prime} } ) \quad 0 < C_i^\ast < 1$$

$C_i^\ast$ Value closest to 1 is selected.

This particular study employs a hybrid model based on a combination of AHP-TOPSIS. The use of such hybrid models is not new in the literature. For instance, Berdie et al. (Reference Berdie, Osaci, Muscalagiu and Barz2017) used the AHP-TOPSIS hybrid model to select the best-integrated software based on customer requirements, project maintenance and time consumption. However, it has not been applied in farm choice decisions earlier.

Study area, study design and data

Study design

The study employs an MCDM based hybrid model using AHP-TOPSIS. The study design is described in Figure 1. After initially preparing the objectives and study of scholarly literature, a survey was undertaken (details of the survey are placed in the next section). This survey collected information from the farmers on their choice between organic and inorganic farming and factors that influence their decision to choose between farming types. These factors are listed in Figure 2. Once this information is collected, expert opinion was soughed through another survey to rank various factors used in this study. Finally, AHP and TOPSIS were applied to select the best alternative.

Fig. 1.

Overview of research process.

Fig. 2.

List of different attribute.