The environmental impacts of economic activities have become an of import facet in the rating of their overall public presentation. Systematically, traditional steps of proficient efficiency have been extended to measure both the economic and environmental public presentation of houses. This survey aims to utilize these steps to analyze a sample of Catalan cultivable harvest farms. Specifically, we apply the methodological analysis late developed by Coelli et Al. ( 2007 ) and widen it to a consideration of the stochastic conditions under which production takes topographic point as proposed by Chambers and Quiggin ( 1998 and 2000 ) . Result suggest that sample farms reach proficient and environmental efficiency degrees on the order of 93 % and 74 % , severally.
Keywords: Environmental and proficient efficiency, Materials balance status, State-contingent engineerings.
JEL Classification: C61, D24, Q12.
Intensive agricultural systems have several harmful impacts on worlds, animate beings and the environment. This has increased societal and political concerns sing agriculture-related negative outwardnesss. At the political degree, these concerns have led European Union ‘s ( EU ) agricultural policies to progressively concentrate on environmental considerations. Interest in advancing agricultural patterns that minimize pollution has been turning. Systematically, different policies have been devised to promote husbandmans to utilize less chemical inputs and to follow environmentally friendly patterns. These alternate patterns nevertheless, can impact the productiveness and efficiency with which farms are runing, which in bend can act upon their economic viability.
Since its origin, the EU ‘s Common Agricultural Policy ( CAP ) has been continuously reshaping itself. While initial aims focused on farm income support, policy Scopess have been widened to encompass environmental saving. Late CAP reform proposals made by the European Commission purpose at alining the CAP with the marks of the “ Europe 2020 model ” and name for environmental sustainability, higher efficiency, effectivity and equitability. Noteworthy among the reform proposals is the purpose to redistribute CAP direct payments on the footing of both economic and environmental standards. In visible radiation of current CAP reform debates, it is of import to develop tools to back up supervising the impacts of policy and to help in better targeting policy steps.
Derivation of farm-level proficient and environmental efficiency ( TE and EE, severally ) indices should be a relevant tool for improved CAP payment redistribution. While high TE steps are a pre-requisite for economic sustainability, high environmental public presentation indexs should lend to environmental sustainability of agricultural patterns. Recent literature on efficiency has been debating on the equal methods to deduce these steps.
Farm-level passage to environmentally sustainable patterns can be regarded as a three-stage procedure affecting different grades of environmental impact decrease. The three phases are efficiency, permutation and redesign ( Wossink and Denaux, 2007 ) . Our analysis focuses on the first stage, which aims at minimising the usage of fouling inputs and optimising input allotment to accomplish the coveted end product degrees. A farm can be considered as environmentally inefficient, if pollution per unit of input is above an ideal lower limit. On the other manus, proficient inefficiencies arise when houses are unable to maximise their end product degrees with minimal usage of inputs ( Farrell, 1957 ) . While privately-run farms are likely to be instead efficient sing conventional input/output usage, they are less likely to be environmentally efficient due to a deficiency of economic inducements and information, bounded reason, or deficiency of external competitory force per unit area sing environmental public presentation ( Wossink and Denaux, 2007 ) .
During the last decennaries, the scientific community has produced several research surveies that attempt to measure the aggregative external costs of modern agribusiness ( Pimentel et al. , 1992 and 1995 ; Evans, 1995 and 1996 ; Bailey et al. , 1999 ; Tiezzi, 1999 ; Pretty et al. , 2000 ; Le Goffe, 2000 ) . Tegtmeier and Duffy ( 2004 ) place the value of the negative impact of agribusiness on H2O, land, air and human wellness around 29.44 – 95.68 dollars per hectare in the USA. In another survey, Pretty et Al. ( 2000 ) obtained a greater value of 363 dollars for the United Kingdom.
In contrast to most sustainability indexs that have been defined at an aggregative degree, farm-level efficiency steps are straight linked to tauten direction determination devising. While aggregate-level surveies can be really utile for politicians and society at big, and can assist planing suited agricultural and environmental policies, they do non supply utile information for determination doing units ( DMU ) who are more concerned about the economic and environmental public presentation of their retentions. Therefore, unlike many environmental public presentation steps that have been defined at the aggregative degree, our survey focuses on gauging combined steps of TE and EE at the microeconomic degree.
The paper is organized as follows. In the following subdivision, a literature reappraisal and the part of this work to old literature is presented. Then, we describe the methodological analysis used in our empirical analysis. The 4th subdivision presents the informations and consequences from the empirical execution. We finish the paper with reasoning comments.
The first efforts to mensurate firm-level EE considered a house ‘s environmental impacts either as an input or an end product. Fare et Al. ( 1989 ) , utilizing Data Envelopment Analysis ( DEA ) techniques, incorporated pollution into productive efficiency analyses as a decrepit disposable bad end product, which implies the premise that the decrease of pollution is expensive. The latter implies that, for a given engineering, cut downing pollution comes at the cost of cut downing the sum of the good end product. The writers developed a inflated productive efficiency theoretical account that treats desirable and unwanted end products unsymmetrically, i.e. , while manufacturers get recognition for supplying desirable end products, they are being penalized for bring forthing unwanted end products. This “ enhanced ” efficiency step was compared with the conventional efficiency index where, by keeping inputs fixed, one expands desirable end products and ignores pollution. The comparing of these two efficiency steps shows the extent to which disregarding unwanted end products distorts the magnitude of efficiency.
Piot-Lepetit and Vermersch ( 1998 ) studied the TE and EE of a sample of Gallic farms specialized in hog production. They considered pollution caused by organic N as an unwanted end product. Using DEA techniques and presuming weak disposability of organic N, the writers found a little ability to cut down N pollution for given end product degrees. Asmild and Hougaard ( 2006 ) used a DEA theoretical account to gauge TE and EE of a sample Danish hog farms. Unlike other surveies that consider pollution as an unwanted end product or input, Asmild and Hougaard ( 2006 ) disaggregated the alimentary excesss into two flows: input and end product. They suggested that EE degrees are around 34-56 % , which shows an of import border to better the environmental public presentation of these farms.
The work carried out by Arandia and Aldanondo ( 2007 ) constitutes an exclusion to published literature on TE and EE of Spanish farms. They focused on a sample of vino farms and studied their Tellurium by agencies of a directional distance map that is fit to informations by utilizing DEA methods. Two bad end products were considered: N and pesticide pollution. Under the strong disposability premise, the mean inefficiency was about 25 % . If the weak disposability was imposed, inefficiency degrees were reduced to 4 % , the difference being caused by the chance costs of pollution decrease, which are higher in conventional agriculture.
Forsund et Al. ( 2008 ) and Murty et Al. ( 2011 ) have shown that reduced-form engineerings that consider pollution either as an input or as a decrepit disposable end product, have serious failings. The stuffs balance rule, that represents the cardinal function of inputs in residuary coevals, is proposed as an appropriate method to pattern pollution. Based on this rule, Coelli et Al. ( 2007 ) suggest a new attack which, in contrast to old research, does non necessitate the debut of an excess pollution variable in the production theoretical account. Coelli et Al. ‘s ( 2007 ) illustrate their proposal by analyzing the environmental public presentation of a sample of farms specialising in pig-finishing in Belgium utilizing DEA techniques. Consequences suggest the being of an ample border to cut down pollution in a cost-reducing manner.
In line with mainstream efficiency analysis, Coelli et Al. ( 2007 ) rely on the premise that the implicit in production engineering is deterministic. When the stochastic belongings of production is ignored, bad production results due to stochastic factors can be erroneously considered as production inefficiencies ( Chambers and Quiggin, 2000 ; O’Donnell et al. , 2010 ) . The state-contingent attack proposed by Chambers and Quiggin ( 1998 and 2000 ) and built upon the theory developed by Debreu ( 1959 ) , is based on the premise that production under uncertainness can be represented by distinguishing end products harmonizing to the province of nature in which they are realized. This leads to a stochastic engineering based on a state-contingent input correspondence.
Under conventional representations of stochastic engineerings, input-output relationships are studied conditional on the realized province of nature. As a consequence, non-substitutability between state-contingent end products is imposed, i.e, it is assumed that manufacturers can merely react to random dazes by modifying input usage, but non by re-allocating state-contingent end products. This representation is known as the output-cubical engineering ( Chambers and Quiggin, 2000 ) , and has been shown to potentially take to of import prejudices in efficiency estimations ( O’Donell et al. , 2010 ) . Ex-ante steps of the random variables are required to get the better of the output-cubicality premise. Since historically, attempts have been focused on roll uping ex-post production informations, ex-ante steps are normally unavailable. When such steps are on manus, the methods used to analyze deterministic engineerings can be easy applied to measure state-contingent engineerings.
Despite the relevancy of state-contingent techniques, there are really few empirical applications based on this methodological analysis. O’Donnell and Griffiths ( 2006 ) usage Bayesian techniques to gauge a state-contingent production frontier. Chavas ( 2008 ) develops a state-contingent cost map and an econometric procedure to retrieve the ex-ante engineering from the ex-post production informations. Following Chavas ( 2008 ) , Serra et Al. ( 2010 ) use state-contingent techniques to measure production determinations in US agribusiness over the last century.
Very few proposals to mensurate EE based on the stuffs balance rule have been presented, which opens the door to important literature parts. Furthermore, none of these surveies has explicitly used state-contingent methods, proposing a clear extension way in this field. The aim of this survey is to deduce combined proficient and environmental efficiency indices for a sample of Catalan farms specialized in cultivable harvest production. In contrast to most environmental public presentation steps that have been defined at the aggregative degree, single efficiency tonss are estimated for each farm in the sample. To accomplish this nonsubjective Coelli et Al. ‘s ( 2007 ) efficiency steps are extended to a consideration of the stochastic environment in which production takes topographic point.
Recently, traditional steps of TE have been extended to incorporate pollution considerations. Late developments within this literature have stressed the necessity to see the stuffs balance status in order to supply sound steps of farms ‘ environmental public presentation. Coelli et Al. ‘s ( 2007 ) proposal, based on this rule, is extended to let for the stochastic conditions of production.
See a house that uses a vector of inputs, to bring forth a province contingent end product, . The set of provinces of nature is represented by, and represents the end product realized under province of nature. The executable production set under the state-contingent attack, , can be derived as follows:
, ( 1 )
where the production engineering is defined to be bulging, non-increasing in inputs, non-decreasing in end products, and strongly disposable in inputs and end products. Let stand for a excess step calculated utilizing a stuffs balance equation, which is specified as a additive map of input and end product vectors:
, ( 2 )
where and represent vectors of known non-negative invariables. The optimisation job seeks to find the optimum combination of inputs for a given sum of end product that minimizes the sum of the excess ( pollution caused in the production procedure ) .
Under the premise that the end product vector is fixed or that the end product vector is non capable of change overing the polluting input into a useable signifier ( i.e. , equals the zero vector ) , the first constituent of the excess in equation ( 2 ) will be minimized when the sum pollutant content of inputs is minimized. For a given vector of pollution contents, , the minimal pollution associated with bring forthing a specified sum of end product can be expressed as:
, ( 3 )
Denote the solution to the minimisation job in ( 3 ) . and stand for the lower limit and the ascertained environmental harm, severally. Following Farrell ( 1957 ) , the technically efficient input vector can be determined by work outing the undermentioned optimisation job:
, ( 4 )
where is a scalar that takes a value between zero and one. is determined by and the corresponding sum of pollution can be approximated by. Following Fare et Al. ( 1994 ) , the DEA linear programming theoretical account to measure input-oriented TE degrees can be expressed as:
( 5 )
where represents the figure of farms. The restraint is included to let for variable returns to scale ( VRS ) .[ 1 ]The EE step proposed by Coelli et Al. ( 2007 ) is expressed as a ratio of minimal pollution over ascertained pollution:
( 6 )
where takes a value between zero and one, the latter indicating that the house is to the full environmentally efficient. The EE tonss for observation, utilizing the DEA method ( Coelli et al. , 2007 ) , are obtained from the following minimisation job:
( 7 )
Harmonizing to Coelli et Al. ‘s ( 2007 ) theoretical account, environmental inefficiencies are caused both by proficient inefficiencies that imply an inordinate usage of fouling inputs, and by allocative inefficiencies affecting an inappropriate input blend given the ascertained vector. Hence, EE is decomposed into two constituents: Tellurium and environmental allocative efficiency ( EAE ) :
, ( 8 )
, ( 9 )
All these efficiency steps ( TE, EE, and EAE ) take a value runing from nothing to one and can be related through the undermentioned look:
. ( 10 )
Though a generalisation of research consequences is hard to do and informations and methodological analysiss used by different analyses are instead heterogenous, amendss derived from pesticide usage are found to be one of the most relevant agriculture-related outwardnesss. This survey focuses on pollution derived from pesticide usage. In line with Morse et Al. ( 2006 ) , Wossink and Denaux ( 2007 ) and De Koeijer et Al. ( 2002 ) , we build an index of pesticide taint that accounts for the sum of pesticide applied and its toxicity. Through a farm-level study, we collected elaborate information on the measures of active ingredients applied through weedkillers, antifungals and insect powders. The entire measure of active ingredients, expressed in litres, was considered as a polluting input. As good known, different active ingredients have different environmental and wellness effects. To deduce a individual step of pesticide pollution, we used a weighting process.
While different indices have been elaborated to mensurate the impacts of different active ingredients on the environment, animate being and human wellness, they normally focus on a limited list of active ingredients. The Acceptable Daily Intake ( adi ) , obtained from the Footprint ( 2012 ) dataset, is the lone index covering the full scope of active ingredients used by our sample farms. The adi measures the measure of active ingredients that can be daily ingested over a life-time, without connoting a important wellness hazard for worlds. This index is normally measured in milligram per kg of organic structure weight per twenty-four hours ( mg/kgbw/day ) . The vector of n adis associated to the active ingredients identified in our sample can be represented as:
, ( 11 )
the vector of weights applied to each active ingredient is expressed as:
.[ 2 ]( 12 )
Pesticide, insect powder and weedkiller pollution is approximated by: , being the measure of active ingredient I applied.
Datas and Consequences
Our analysis uses transverse sectional, farm-level informations collected utilizing a questionnaire that was distributed among 190 agricultural retentions specialized in the production of cereals, oil-rich seeds and protein ( COP ) harvests. The study was conducted during the planting season in 2011 and includes inside informations on planned input usage, fiscal, socio-economic features and structural features of sample farms. We besides obtained informations on ex-ante end products for three alternate provinces of nature: bad, normal and ideal turning conditions.
End product represents farm COP production in Euros ( a‚¬ ) under province. Inputs included in the analysis are: entire hectares ( hour angle ) of land planted to COP ; hired and household labour, expressed in hours ; capital input that aggregates the replacing value ( in a‚¬ ) of machinery, other equipment and edifices used in the production procedure ; the outgo in fertilisers expressed in a‚¬ ; pesticide, weedkiller ad insect powder usage ( litres of active ingredients ) ; seed disbursals, expressed in a‚¬ ; energy input including fuels and lubricators, expressed in a‚¬ ; contract work, expressed in a‚¬ .
Table 1 provides drumhead statistics for the variables used in the analysis. Sample farms cultivate, on norm, 75 hectares, a farm size above the national and the EU cultivable harvest farms ‘ norm, of around 52 and 46 hectares, severally ( Farm Accountancy Data Network, FADN 2012 ) . More than 95 % of the COP country is devoted to wheat ( 36 % ) and barley ( 60 % ) production. Sample farms devote, on norm, 552 hours to COP production during the turning season, of which more than 90 % represents unpaid household labour.
Depending turning conditions, husbandmans expect to obtain different end product degrees. While under bad conditions the mean value of COP production is around 31 thousand a‚¬ per farm, under ideal conditions mean end product is on the order of 70 thousand a‚¬ . Under normal conditions, the value of COP production ( 51 thousand a‚¬ ) generated by our sample farms about doubles the EU farms ‘ norm end product ( 28 thousand a‚¬ ) . Per hour angle statistics suggest our sample farms are more intensive than both national and EU farms: while EU and national farms have severally, an mean income of 627 and 428 a‚¬ per hour angle, 670 a‚¬ per hour angle is the mean income of our sample farms under normal turning conditions.
Machinery and edifices used by sample farms amount to 134 1000 a‚¬ , or 2,304 a‚¬ per hour angle, above EU ‘s mean investing ratios on the order of 1,497 a‚¬ per hour angle ( FADN, 2012 ) and it is much higher than the national norm ( 536 a‚¬ per hour angle ) . While sample farms cultivate more land than EU cultivable harvest farms, they spend less money in fertiliser than the latter ( 5,315 vs. 9,279 a‚¬ yearly ) . Entire measure of active ingredients used by our sample farms is, on norm, on the order of 85 litres. Expenses in pesticides sum to 2,975a‚¬ , below EU ‘s mean harvest farms ( 5,120a‚¬ ) . Annual disbursals in seeds and energy are on the order of 3,866 and 4,913 a‚¬ , severally.
The DEA analysis efficiency consequences are presented in table 2. Result show average TE tonss on the order of 0.93 and 0.87 under VRS and changeless returns to scale ( CRS ) , severally, proposing that our sample farms could utilize on mean 7 % ( 13 % ) fewer inputs to bring forth the same degree of their current end product. Tax returns to scale were studied to happen that farms operate under increasing returns to scale. More than 70 % of the observations have efficiency evaluations greater than 0.90, demoing comparatively high public presentation degrees. However, under CRS, more than one half of farms exhibit this public presentation. Under VRS ( CRS ) premise, proficient efficiencies range from a lower limit of 0.565 ( 0.183 ) to a upper limit of 1, proposing that our sample husbandmans present different accomplishments to pull off their retentions.
Serra and Goodwin ( 2009 ) , who used a local upper limit likeliness attack, found Spanish conventional ( organic ) cultivable harvest farms to be characterized by average efficiency tonss of 0.97 ( 0.94 ) , values that are comparatively close to our findings for VRS. Our sample farms show better public presentation than the sample of Catalan grape farms studied by Guesmi et Al. ( 2011 ) , utilizing a stochastic frontier attack. The writers obtained proficient efficiency tonss of 0.80 and 0.64 for organic and conventional grape farms, severally. Our sample farms besides exhibit higher proficient efficiency tonss than Coelli et Al ‘s ( 2007 ) consequences for a sample of Belgian hog coating farms ( 0.929 vs. 0.897 ) . Ignoring the stochastic nature of production normally consequences in overestimate of inefficiency steps ( Chambers and Quiggin, 2000 ) .
The mean EE tonss are 0.737 ( 0.577 ) under VRS ( CRS ) premise, bespeaking that husbandmans should be able to bring forth their current end product with 26 % ( 42 % ) fewer pesticide, weedkiller and insecticide pollution. Result suggest that husbandmans who are environmentally efficient tend to be more technically efficient, back uping that an efficient usage of chemical inputs improves both environmental and proficient public presentation. Environmental efficiencies range from a lower limit of 0.02 to a upper limit of 1, proposing of import scattering within Catalan farms.
Previous research, utilizing different methodological analysiss and/or measuring different instance surveies has derived different consequences. Coelli et Al. ( 2007 ) obtained higher EE evaluations on the order of 0.843 for his sample of pig-finishing farms in Belgium. Asmild and Hougaard ( 2006 ) who assessed the EE of a sample Danish hog farms, obtained much lower EE degrees ( 34-56 % ) .
A relevant part of environmental inefficiency is due to EAE. Sample farms have, on norm, an EAE mark of 0.758 under VRS and 0.606 under CRS. Sample farms, therefore, do non hold the right input mix, given the observed a vector. Bettering proficient and environmental efficiency degrees can cut down production costs and better the economic viability of farms. It can besides vouch higher environmental quality.
The productive efficiency literature has paid really small attending to environmental public presentation issues. Turning societal and political concerns for the environmental impacts of agribusiness make it necessary to analyze environmental and proficient public presentation utilizing robust methodological analysiss that enable to deduce dependable public presentation indexs. Recently, Coelli et Al. ( 2007 ) proposed a new attack based on the stuffs balance construct that represents the relevant function of inputs in bring forthing remainders.
This survey contributes to the literature by widening Coelli et Al. ‘s ( 2007 ) proposal to let for the stochastic environment in which production takes topographic point. The extension is based on the state-contingent methods ( Chambers and Quiggin, 2000 ) . The theoretical account is applied to deduce combined proficient and environmental efficiency degrees achieved by 190 Catalan farms specialized in cereals, oil-rich seeds and protein harvest production. To our cognition, this is the first empirical application that derives TE and EE steps utilizing both the stuffs balance and state-contingent models. Our empirical findings suggest that our sample farms, on norm, make proficient efficiency tonss of 93 % and therefore that they can cut down input usage by 7 % while go forthing end product degrees unaltered. The mean environmental efficiency mark, on the order of 74 % , indicates ample range to better environmental public presentation and cut down pesticide usage and pollution by 26 % . These inefficiencies are, to a big extent, caused by allocative inefficiencies.
Some policy recommendations to increase the comparatively low EE degrees are as follows. First, since chemical input is partially applied out of wont ( husbandmans tend to make what they have done in the yesteryear ) . Information and preparation classs on how to adequately use chemical inputs may better the agricultural sector ‘s environmental public presentation. Second, CAP subsidy redistribution on the footing of environmental standards, may move as an effectual tool to actuate husbandmans to follow environmentally friendly patterns. Finally, since environmental inefficiencies are chiefly due to allocative issues, supplying husbandmans with better information on the environmental impacts of different pesticides, weedkillers and insect powders, should better environmental public presentation. Economic incentives punishing the usage of those chemicals with stronger harmful effects should farther guarantee better environmental efficiency degrees.