Next Article in Journal
A Three-Phase Four-Leg Inverter-Based Active Power Filter for Unbalanced Current Compensation Using a Petri Probabilistic Fuzzy Neural Network
Previous Article in Journal
Large Signal Stabilization of Hybrid AC/DC Micro-Grids Using Nonlinear Robust Controller
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Testing for Environmental Kuznets Curve in the EU Agricultural Sector through an Eco-(in)Efficiency Index

1
Department of Agriculture Crop Production and Rural Environment, University of Thessaly, Fytoko, 38446 Volos, Greece
2
Department of Planning and Regional Development, University of Thessaly, Pedion Areos, 38334 Volos, Greece
3
Department of Industrial and Systems Engineering, University of Florida, 401 Weil Hall, Gainesville, FL 32611-6595, USA
*
Author to whom correspondence should be addressed.
Energies 2017, 10(12), 1992; https://doi.org/10.3390/en10121992
Submission received: 1 October 2017 / Revised: 27 October 2017 / Accepted: 26 November 2017 / Published: 1 December 2017

Abstract

:
Studies on agricultural production practices advance within international literature and new methods are proposed in order to assess the agricultural sustainability, either at farm or macro level. The present paper builds on these advancements and develops a synthetic Eco-(in)efficiency index by employing a directional distance function—data envelopment analysis (DEA) model. This index is used in order to assess the sustainability of the EU agricultural sector for the period 1999–2012 on a country level. Furthermore, Eco-(in)efficiency, together with the energy use and greenhouse gas (GHG) emissions are regressed on the gross domestic product (GDP) of EU countries, in order to check for any environmental Kuznets curve relationship existence. Results signify that efficiency improvements are possible, both towards output development and GHG emissions reduction. In addition, the potential of each country in adopting more sustainable production practices is not totally connected with its economic development, as Eco-(in)efficiency and GDP levels of EU countries seem to be linked with an N-Shaped curve.

1. Introduction

Uncontrolled economic growth imposes severe risks for the environment as natural resources used for economic activities are not infinite [1]. Ice melting and the increase of temperature are considered as tangible examples of environmental degradation due to human activities [2]. After a long period of uncontrolled economic growth, policy makers, mostly representing matured economies, started to realize the risk imposed because of environmental degradation and initiated global actions in order to set the basis for sustainable development. Undoubtedly, the benchmark point towards sustainability has been the United Nations (UN) World Conference on the Environment which was held in Stockholm in 1972. Although the debate did not succeed in a strong commitment towards more environmental-friendly economic growth, its 26 common-agreed principles regarding environment and development could be regarded as the first common achievement towards global sustainability [3,4].
In the following years, a heated debate around the limits between environmental protection and economic growth evolved across academics, policy makers and practitioners. A common belief is that environmental degradation increases during the early stages of economic development. After a certain point of growth the trend is reversed, in a sense that development becomes more sustainable [5]. The rationale behind this trend is that in early stages of development, pollution-generating economic sectors are developed without concrete control, since economic targets surpass the respective environmental ones, and environmental degradation is an accepted byproduct of the desired economic outcomes. In addition, as income rises most people start to put a larger weight on their quality of life and thus environmental issues are addressed by legislation in a more systematic context [1].
This kind of functional relationship between environmental degradation and economic growth is graphically expressed by an inverted U-shape slope called the environmental Kuznets curve (EKC) [6]. Many scholars have tried to empirically verify the existence of EKC for different environmental dimensions (i.e., CO2 emissions, particulate matter, energy use) and in various geographical and time contexts. The papers of [7,8] provide two comprehensive and detailed reviews of the relevant past empirical studies.
As far as the agricultural sector is concerned, production comes up with several environmental damages that mainly have to do with land–take, habitat loss, soil erosion and degradation, water contamination and climate change [9]. Nevertheless, research on the relationship between negative environmental impacts and economic growth is still limited. More precisely, the majority of studies focus their interest on developing countries seeking empirical evidence on the relationship between economic development and agricultural land expansion, particularly against forests [10,11]. Besides these studies, to the best of our knowledge, there is only one study examining environmental risks of agriculture in advanced economies. More precisely, ref. [12] analyzed the longitudinal fluctuations of environmental risks arising from pesticides used in agriculture within US states against real gross state product, abatement efforts and an environmental productivity index. Managi [12] argued that although environmental risks and economic growth seem to validate the EKC relationship, the risk of observing an N relationship still exists due to scale effects. Thus, even for advanced economies, such as USA, EKC in agriculture could not be taken for granted.
The existing literature testify that, for agriculture, there is a lack of studies regarding advanced economies’ dynamics, such as those of EU countries, and other environmental impacts such as GHG emissions remain rather unexplored. The importance of examining the atmospheric impacts of agriculture is extracted from the fact that global agriculture accounts for about 24% of the total GHG emissions. In addition, the Organisation for Economic Co-operation and Development (OECD) countries account for more than one third of global emissions [13]. In a European context, the share of agriculture in total GHG emissions is exceeding 10%, with a total production of 470.6 million tons of CO2 in 2012. Nevertheless, the contributions vary significantly across countries. To illustrate, Irish, Latvian and Lithuanian agricultural sectors account for more than 20% of the total national emissions, whereas the share of agricultural GHG emissions in countries such as Malta, Luxemburg and the Czech Republic do not exceed 7% [14]. In addition, focusing on agricultural nutrients, EU countries have managed to reduce their nitrogen and phosphorus balances during the period 2000–2012. Nitrogen Balance for EU-28 has been reduced from 63 to 51 and phosphorus from six to two. Nevertheless, efficiency in nutrient utilization, as this is expressed by the ratio of total nutrient outputs to total nutrient inputs, also varies profoundly across countries. For the period 2009–2014, the most efficient countries in nitrogen efficiency was Romania, Ireland and Bulgaria and the least efficient ones were Cyprus, Luxemburg, Greece, Malta and Croatia [15].
Undoubtedly, the variations among EU countries reflect, in a straightforward context, the composite nature of agricultural production sustainability issue. Countries such as Malta and Luxemburg present positive figures regarding GHG emissions whilst being inefficient in nutrient utilization. The opposite stands for countries such as Ireland. Such variations have been also identified by the recent study of [16] who employed the generalized Divisia Index to evaluate the effect of energy use and agricultural growth in energy-related GHG emissions of 17 EU countries from 1995 to 2012. These variations portray, that in order to extract a comprehensive picture regarding the sustainability levels of national agricultural practices, the analysis should focus on all aspects of agricultural production. That is to say, not only the amount of inputs or the produced outputs could define sustainability, but the level of efficiency of how inputs are utilized in order to produce certain outputs that lead to a comprehensive evaluation index of agricultural sustainability [16].
The link between sustainability and efficiency is reflected in the recent reforms of the common agricultural policy (CAP) which precisely focus on increasing the efficiency of use of agricultural inputs, and on reducing the production of undesirable outputs, such as GHG emissions [17]. Nevertheless, this goal requires the development of various technologies on both the production and application of agricultural inputs, like fertilizers, agrochemicals, water and fuels [18]. Additionally, efficiency of inputs use is an integral part of sustainability where economic and environmental parameters merge, in order for production cost reduction and mitigation of contamination to occur [19]. It is obvious that the economic and the environmental performance of agricultural production are controversial parameters, being at the core of a continuous debate between farmers and various stakeholder groups, with the latter requiring from the former to adjust the production process towards an environmentally friendly approach [20]. The experience of other economic sectors has shown that the costly environmental-friendly amendments to production are more easily applied in advanced economies [21,22]. By extension, and regarding the agriculture sector, mature economies are expected to present more sustainable agricultural production practices, such as those having resolved basic dietary issues, are usually characterized by increased environmental awareness and their demand focus on safe and clean, in environmental terms, agricultural products.
The justification of this debate necessitates the need for empirical evidence. As the relationship between agricultural sustainability and economic development is still rather unexplored by international literature, the present paper seeks to fill this gap by exploring the potential existence of an EKC for the European agricultural sector. In order to account for all dimensions of agricultural sustainability, a synthetic Eco-(in)efficiency indicator is developed and applied to the 28 EU countries in order to evaluate sustainability variations for the period 1999–2012. At a later stage, the countries’ figures are regressed against GDP p.c. in order to test for an EKC relationship. The remainder of the paper is as follows. Section 2 describes the composition of the eco-efficiency indicator and the regression details for testing for the EKC function. In Section 3, the variables employed by the present study are explained and the results of the analysis are presented and discussed. The paper ends up with the main conclusions and some policy implications towards more sustainable agricultural practices on the European level.

2. Materials and Methods

2.1. Defining the Eco-(In)Efficiency Index

2.1.1. Considerations Regarding DEA Eco-Efficiency Indicators Development

The challenge of conceptualizing and measuring sustainability has captured the interest of scholars across various disciplines in developing synthetic quantitative indices which could be used as the basis of sustainability evaluation [23]. Towards this direction, eco-efficiency could be regarded as an effective approximation of sustainability since it measures the efficiency of a production unit in maximizing its desirable outputs whilst keeping the production of its undesirable outputs and the use of natural resources at the lowest possible levels [24,25]. That is, eco-efficiency could produce integrated assessments of various dimensions of production efficiency such as technical, economic, ecological, energy and environmental [26].
A large part of literature is based on production frontier theory, firstly introduced by [27], in order to develop eco-efficiency indices. The main methodologies based on the frontier theory are data envelopment analysis (DEA) and stochastic frontier analysis (SFA) [25,28]. The two methods are based on different assumptions concerning their application. DEA is a non-parametric method which involves the solution of a series of linear programming problems in order to extract the relative efficiency of each Decision Making Unit (DMU). In addition, SFA is based on the specification of a production function and efficiency estimates are extracted with the use of statistical regression models [29]. Between the two methods, DEA could be regarded as more easily applicable and thus has been rendered rather popular towards eco-efficiency estimations [30].
Among the various specifications of eco-efficiency DEA models there are some basic issues that should be taken into account before its empirical application. The first refers to the reference set and the type of desirable and undesirable output relationship that would be chosen for the formulation of the model. On the one hand, there are studies that assume weak disposability among outputs meaning that reductions of undesirable outputs result in desirable production losses [31,32]. On the other, when strong disposability is assumed, then reductions of undesirable outputs could be achieved without any reduction in the desirable outputs [33]. In addition, another issue that should be defined is whether the models are based on simple distance functions or directional distance functions. In the first case of studies, the adjustments of desirable and undesirable outputs are headed to the same direction after all undesirable outputs have been transformed mainly by using their reciprocals [34,35,36], whilst in the second case these could be realized to opposite directions [37,38,39]. In addition, before the formulation of a DEA model, one should also consider if the adjustments of inputs or outputs would be proportional or non-proportional. In the first case, radial DEA models are used [40,41] and in the latter, non-radial measures are employed [42,43]. The papers of [44,45] deal extensively with the aforementioned issues providing comprehensive reviews regarding the application of the DEA method in sustainability assessments taking into account the energy and environmental dimensions of production.

2.1.2. DEA Applications on Agricultural Sector

As far as agricultural studies are concerned, the latest relevant literature has shown a remarkable progress in developing synthetic methods towards sustainability assessment, lying far from a simple measurement of output production or input use [12,13,16]. More precisely, until recently, most of the attention was given to the technical efficiency of agricultural sectors. In a cross-country context, ref. [46] assessed the technical efficiency of 93 countries for the 1980–2000 period based on DEA, ref. [47] analyzed the impact of CAP subsidies on the technical efficiency of crop farms in Germany, the Netherlands and Sweden and [29] assessed the longitudinal technical efficiency of EU countries by comparing the results of DEA and SFA methods. Martinho [48] employed DEA to assess the efficiency and total factor productivity in the European agricultural sector at both farm and regional level. Finally, ref. [49] extracted technical efficiency scores for EU countries and used a bootstrap method in order to extract their statistical properties.
In recent years, the notion of efficiency estimation through frontier deviations has been extended to account for environmental and energy dimensions of agricultural production. The flexibility of DEA provided a fertile ground for such modifications. In this second category of agricultural studies, it was [50] who firstly decomposed agricultural efficiency in its technical, economic and environmental dimensions. In addition, ref. [26] assessed the sustainability of Dutch sugar beet growers by taking into account indices of technical efficiency, environmental efficiency and a composite index of sustainable efficiency. Managi [12] modified the total factor productivity Malmquist–Luenberger index, in order to account for environmental factors. Later, ref. [51] developed a composite sustainability indicator consisting of the dimensions of technical efficiency and cumulative exergy allocative efficiency and evaluated the sustainability of the agricultural sector of OECD countries. Picazo-Tadeo et al. [24] extended the basic DEA model and used a directional distance function (DDF) in order to extract different estimates of the efficiency of olive-growing farms in Spain. Since both undesirable inputs and outputs were considered, authors regarded these estimations as representing the eco-efficiency of the DMUs’ sample. Vlontzos et al. [30] employed a non-radial DEA model in order to extract different estimations of the environmental and energy efficiency of EU countries. Moreover, Tian et al. [52] used a slacks-based DEA model in order to incorporate undesirable agricultural emissions in the efficiency evaluation of rice farmers of China. The slacks of energy inputs have been also assessed by [53] within the technical efficiency measurement of French farms. Finally, Vlontzos et al. [35] used Window DEA to assess the longitudinal technical efficiency of EU countries taking into consideration their GHG emissions. GHG emissions factor was entered into estimations with two ways. The first involved its consideration as input and the second as output after being rescaled.
The preceding studies provide useful insights into the efficiency of the agricultural sector. Since the aim of this paper is to test for an EKC relationship between agricultural sector and economic growth, estimations of pure technical efficiency could not be used as the basis of the comparisons, as these do not encompass the environmental dimension of production. Thus, the present paper will mostly focus on the models of the second category of studies which can provide a more holistic assessment of agricultural sustainability. These kinds of models have been extensively used for testifying an EKC curve in other disciplines [33,54,55] but have never been adjusted for the agricultural sector. Keeping in mind, that for an EU cross-country evaluation context both non-radial [23] and radial [35] models have been used to quantify environmental efficiency, the present paper follows the rationale of [12,32] in order to enrich the methodological portfolio regarding environmental efficiency in EU countries. It should be noted that the difference between the present study and the study of [12] lies in the scope of efficiency estimation. On this, efficiency in the present paper is estimated in order to be used as a dependent variable in EKC testing, whereas within the context of [12], this was used as an independent variable capturing the environmental technology differences across US states. Taking into account the preceding remarks, the contribution of the present paper is twofold. Initially, it extends the previous sustainability assessments in the agricultural sector with a novel composite DEA indicator based on the DDF rationale. Secondly, it uses this indicator to test, for the first time, for an EKC relationship of agriculture sector in a group of advanced economies such as that of EU members. By doing so, the paper enriches the agricultural literature but also contributes to the global debate around the relationship between environmental degradation and economic growth.

2.1.3. Developing the DEA Eco-(In)Efficiency Indicator for EU Agricultural Sector Sustainability Assessment

In contrast to the basic DEA formulations which seek to maximize outputs or minimize inputs, DDF allows for simultaneous adjustments on both inputs and outputs. This rationale has been also incorporated in order to provide estimations of productive (in)efficiency when both desirable and undesirable outputs are taken into consideration [32].
Given the presence of x inputs, g desirable outputs and b undesirable outputs, the technology set is defined as:
[ ( x , g , b ; e c o E f f ) : x   can   produce   ( g , b ) ]
In addition, all the feasible output vectors sets for a given input vector are the following:
P ( x ) = [ ( g , b ) : ( x , g , b ) T ]
Then, the directional distance function has the following form:
D T ( x , g , b ; e c o E f f ) = sup δ : ( g + δ e c o E f f g , b δ e f f b ) P ( x + δ e f f x )
Moreover, we suppose that we have n national agriculture sectors expressed as N A S j   ( j = 1 , 2 , , n ) , and x i j > 0 is the amount of input i used in total by National Agricultural Sector (NAS) N A S j , g r j > 0 is the amount of desirable output r of N A S j and b c j > 0 the amount of undesirable output c of N A S j . Then, the inefficiency score for the N A S i is extracted by solving the following model [56]:
max E f f , λ e c o E f f s . t . j = 1 n x i j λ j + e c o E f f i o x x i o i = 1 , 2 , , m j = 1 n b c j λ j + e c o E f f d c o b = b c o c = 1 , 2 , , k j = 1 n g r j λ j e c o E f f d r o g y r o r = 1 , 2 , , d j = 1 n λ j = 1 λ j 0 j   =   1 , 2 , , n d = ( 0 , g , b )
The objective of the model is to maximize the desirable outputs and minimize the undesirable outputs whilst keeping the inputs stable. The value of e c o E f f portrays the Eco-(in)efficiency, meaning that for a fully efficient NAS, the value of e c o E f f will be 0. The constraints of the model portray that a weak disposable production process has been selected for the present dataset, which assumes that reductions in CO2 emissions could not be realized without desirable output losses. This kind of model formulation seems more realistic than a strong disposable approach which allows for CO2 emissions reductions without any loss in desirable outputs [51]. In addition, the model assumes variable returns of scale in the production process, thus eliminating the scale effects and resulting in more reliable targets for each NAS.

2.2. Defining the Model for EKC Testing

After extracting the e c o E f f scores for each country, a regression analysis will be conducted in order to check for an EKC relationship. For comparison reasons, three different models will be tested. The first will test for EKC between Eco-(in)efficiency and GDP. The second will focus on energy use and the third on total GHG emissions. The review of previous studies has shown that different modifications of regression models, such as time-series, cross-section and panel-data have been employed in order to check for EKC in other sectors [7]. Among them, panel-data is the most widely used method for acquiring cross-country empirical evidence [57]. The fundamental consideration regarding the adaptation of a panel-data regression on a study is the functional form of the relationship among dependent and independent variables. Three basic forms could be found in the literature. The first is the linear specification, the second is the log-log specification in which both dependent and independent variables are logged before being entered into the model and the third is the semi-log function in which only the dependent or the independent variables are logged. In the present paper, the linear form is preferred, as it leads to a direct approximation of the turning point of the curve.
The second consideration is the choice between fixed and random effects specification of the model. In fixed effects, the intercept is gathered as fixed whereas in random effects specification, it is assumed that it is extracted through the function a i = a ¯ + μ i , where a   ¯ is an unknown parameter and μ i are unknown iid variables. For the selection of the appropriate specification it is advisable to run a Hausman test before running the statistical model [55]. Finally, it should be noted that since DEA estimations portray the relative efficiency of each NAS, it is expected that a cross panel correlation will emerge in the respective panel data model which could result in biased estimations. To overcome this difficulty we employ a regression with panel-corrected standard errors (PCSE) which controls for heteroscedasticity and contemporary correlation of disturbances across panels [58]. The models specifications are given below:
Y i t = a i + β G D P G D P i t + β G D P 2 G D P i t 2 + β G D P 3 G D P i t 3 + ε i t i = 1 , 2 , , N   countries ,   t = 1 , 2 , , T   years   Y = e c o E f f   for   model   2 . 1 ,   Y = E N E R p c   for   model   2 . 2 ,   Y = G h G p c   for   model   2 . 3

3. Results and Discussion

3.1. Eco-(In)Effiiency Assessment Results

In order to depict the production process of EU agricultural sectors, respective inputs and outputs have been selected. The inputs used in this study are expressed as follows: land variable is expressed in total hectares, capital as the fixed capital consumption in monetary terms and labor as the employees’ annual working units (AWU). In addition, total energy cost and the total cost of chemicals and fertilizers used in the production process have been also incorporated into the model as significant contributing factors to the agricultural output of each country. Moreover, output has been segmented to desirable output (total crop and animal output) and undesirable output (total GHG Emissions). The basic descriptive statistics of the input and output variables of model 1 are presented in Table 1.
The data covers the period between 1999 and 2012. In order to extract the longitudinal Eco-(in)efficiency scores of each country, Model 1 is executed 14 times. The main results for each NAS are presented in Table 1, whereas the detailed annual Eco-(in)efficiency scores for each NAS are given in Table A1 of the Appendix A. Regarding Table 1, the mean Eco-(in)efficiency score for each NAS is given in the first column, the years for which each country presented zero (in)efficiency are presented in the second column, whereas the highest and lowest Eco-(in)efficiency score of each country through the whole 1999–2012 period are given in the third and fourth columns, respectively. As can be seen in Table 2, the average Eco-(in)efficiency score for the EU agricultural sector is 0.11. This finding implies that with the given inputs, EU agricultural could produce 11% more output and less GHG emissions. Nevertheless, variations exist among countries. More precisely, only Malta and the Netherlands kept a zero Eco-(in)efficiency for all the considered period, followed by Italy which was found to be rather inefficient only in one year (2005), and France, which was found to be inefficient in three years (1999, 2005, 2006). On the contrary, the least efficient group of countries consists of Lithuania, Slovakia and Finland, whose average Eco-(in)efficiency scores for the period 1999–2012 exceeded 0.3. A second group of inefficient countries is shaped by Latvia, Hungary, the United Kingdom, the Czech Republic and Slovenia with Eco-(in)efficiency scores ranging between 0.2 and 0.3.
Apart from the top-ranked countries, a number of other national sectors achieved remarkable figures in terms of times being evaluated as efficient during the recursive application of the DEA models. Poland, Spain, Luxemburg and Cyprus were found fully efficient for more than ten years and Belgium, Portugal, Germany, Bulgaria and Ireland for exactly ten years. In addition, Lithuania, Slovakia, Finland Hungary and Sweden have not acquired a zero Eco-(in)efficiency score in any of the annual estimations. Moreover, examining the last two columns, it is interesting to highlight the large variability of Latvia figures, as within the period under consideration its highest Eco-(in)efficiency score was 0.634 whilst the lowest 0. Rather high variability is found also for the United Kingdom, Estonia and the Czech Republic. As it is evident, the least efficient countries present a larger range of scores during the 1999–2012 period.
Having examined the individual records of the countries, Figure 1 presents the annual Eco-(in)efficiency estimations for three categories of countries against the respective scores of the whole sample. The first category of countries is composed of countries that joined the EU before 2004, the second of the countries that entered into the EU in 2004 and the third of the three countries that joined the EU after 2007. This division is critical in order to check the performance of countries with reference to the CAP policy guidelines. In general, average EU Eco-(in)efficiency presents variable trends through the considered period. The least efficient scores are observed over the years 1999 and 2005 whereas the most efficient scores are found in years 2007, 2011 and 2012. Among the group of countries, these that have accessed EU in 2004 are characterized with the highest average Eco-(in)efficiency scores during the whole period. It should be noted that average inefficiency of these countries exceeds the average EU inefficiency at all annual estimations. The other two groups of countries present lower inefficiency records, which is reflected in the fact that in any year their mean inefficiency exceeded the EU average. Finally, comparisons between the inefficiency trends of the two groups of countries, could lead to any safe conclusions, as these present high variability. The only remarkable observation is the significant better performance of late-joining countries during the period 2005–2007.

3.2. EKC Testing Results

As it was stated in Section 2, the scores of Eco-(in)efficiency will be used as the basis for fitting an EKC curve on the relationship of agriculture environmental efficiency and economic growth in EU countries. In addition, the existence of an EKC will be also examined for p.c. energy use of agriculture and p.c. GHG emissions. The annual GDP p.c. of EU countries for the years 1999–2012 will be used as the variable to represent the economic growth of each country. In order to achieve an unbiased representation of the economic level of each country, the GDP will be expressed in purchasing power parity (PPP), as this measure allows for more accurate cross-country comparisons and has been employed in several empirical studies regarding EKC [5]. A cubic function has been selected in order to check for any N-shape relationships. The empirical recognition of an EKC or N-shaped pattern is based on the estimations of the regression parameters and, more precisely, on their signs and their statistical significance. In this sense, an EKC curve would be confirmed only under the following signs of estimated parameters β G D P > 0 ,   β G D P 2 < 0 ,   β G D P 3 = 0 . Finally, it should be mentioned that the final sample includes 27 EU countries. Luxemburg data was omitted from estimations because it presents GDP figures that are exceptionally higher than the respective figures of other countries, and thus could lead to biased estimations.
Table 3 presents the results for the three models. As far as the goodness-of-fit tests are concerned, the statistical significance of the Wald test result denotes that all three models are performing better than the respective models with zero parameters. In addition, for models 2.2 and 2.3, the results of the Hausman test indicate that for both models, a random effect specification could be used with the available data. For model 2.1, estimations present statistical significance for all parameters at the 0.01 level, except for the intercept. The estimations for the β G D P and β G D P 3 are positive, whereas the estimation of β G D P 2 is negative. This finding signifies that eco(in)efficiency and economic growth are linked through an N-Shaped curve. More specifically, in early stages of development, agricultural production is rather inefficient with desirable output shortages and undesirable output excesses. As the national income increases, production becomes more efficient in input utilization but after a certain point of income, inefficiency presents increasing trends again. This fact could be attributed to the ability of wealthier countries to afford more undesirable inputs such as fertilizers and energy. Hints for this explanation are provided by the examination of the estimated coefficients of model 2.2. As for model 2.1, all estimations acquire statistical significance except for the constant term. The pattern of parameter sign change is exactly the same as that of model 2.1. This finding provides hints that Eco-(in)efficiency is strongly connected to energy use. Finally, model 2.3 estimations are all statistically significant with the exception of β G D P 3 . Thus, agricultural GHG emissions seem to be connected to economic growth through an EKC relationship.
As far as the turning points are concerned, these are visualized with Figure 2 which presents the original and fitted values of the dependent variables extracted from the three different models, plotted together with the GDP variable. As can be seen from Figure 2a, Eco-(in)efficiency increases up to 11,326€ and starts decreasing until 29,294€. After this point, there are hints that countries return to more unsustainable practices. Nevertheless, inefficiency at this part of GDP levels is still lower than the respective of the lowest GDP levels. The same trends are observed over the energy use (Figure 2b), although with two remarkable differences. The first regards the turning points, which are closer than those of Eco-(in)efficiency (12,428€ and 24,961€) and the second on the fact that after the second turning point, energy use, in some cases, surpasses the respective use at the lowest GDP levels. Finally, GHG emissions and GDP level relationship seems to match with the characteristics of an EKC curve. Emissions are expanding until the point of 12,325€, and afterwards they present declining trends.
The aforementioned findings provide useful insights into the state and evolution of primary sectors in the EU. A typical analysis focusing only on the GHG emissions would provide hints that the CAP fulfills its targets, at least on the emissions side. As confirmed by the present paper, EKC relationship confirms the general impression that emissions is a matter of economic growth that could be solved together with the global target of EU regional and national economic convergence. Nevertheless, the results of Eco-(in)efficiency and energy use analyses which showed that inefficiency and energy use excesses are also observed in countries with higher income levels, add a question mark for whether EU convergence is only a necessary or/and sufficient conditionfor rendering the EU agricultural sectors more sustainable. The findings of the paper, extend the debate of previous papers [12,16] on the necessity of developing more advanced methods on agricultural sustainability assessments, especially in economies where technological, institutional and scale factors are rather complex.

4. Conclusions

This analysis revealed that EU agriculture is quite far from being fully sustainable. The Eco-(in)efficiency index results signifies that the overall EU primary sector could produce 11% more output and less GHG emissions. Inefficiency seems to be higher for the group of countries that accessed the EU in 2004, as for countries such as Slovakia and Lithuania output shortages and GHG emissions excesses, surpass 30%. Moreover, the EKC analysis revealed that despite the fact that GHG emissions seem to be reduced as national incomes increase, a more advanced economic level does not ensure a more sustainable agricultural production, as a whole. This is reflected in the N-Shaped curve linking the Eco-(in)efficiency and energy use levels with the GDP of EU countries during the period of 1999–2012. Thus, a future success of the convergence process within the EU should not be regarded as a decisive factor of agricultural sustainability improvement.
The contradictory results among the three agricultural sustainability dimensions tested by the present paper highlight the challenge of measuring agricultural sustainability in a more holistic and systematic context. To this end, the models of the present paper could be enriched with additional control variables representing particular characteristics of EU countries, such as, their institutional framework, their technological advancement, their specialization in crop and livestock production etc. Moreover, new synthetic indicators for measuring agricultural sustainability could be added to the ones employed by the present paper. In addition, at the policy level, community initiatives such as the European innovation scoreboard should be capitalized and adapted to the specific challenges of the EU agricultural sector.
Finally, further improvement of efficiency scores for inefficient countries can be achieved through utilization of good practices being implemented in efficient ones. It is therefore necessary that advanced technology are reachable by farmers, hence more effectively and efficiently utilizing energy resources, while reducing GHG emissions. The continuous training of farmers is also necessary in order to increase their ability to adopt and implement new cultivation and production practices. It has been proven that continuous engagement with new knowledge leads farmers to more appropriate decision-making processes which have a positive impact on efficiency improvements [62]. The new CAP provides considerable degrees of freedom to farmers to organize their cultivation and production plans, and is an appropriate framework for efficiency change and efficiency improvement.

Author Contributions

All authors designed the structure of the paper and contributed to its writing; George Vlontzos prepared the data, conducted the literature review and estimated the Eco-(in)efficiency scores; Spyros Niavis and Panos Pardalos developed the framework for Eco-(in)efficiency estimation and tested the EKC curve.

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A

Table A1. Annual Eco-(in)efficiency scores of EU agricultural sector (1999–2012).
Table A1. Annual Eco-(in)efficiency scores of EU agricultural sector (1999–2012).
Year
Country19992000200120022003200420052006200720082009201020112012
Belgium0.060.000.000.070.000.000.010.000.000.000.000.000.000.00
Bulgaria0.000.000.000.000.000.000.000.000.000.000.130.030.030.11
Czech Republic0.000.000.000.000.000.390.450.440.270.270.400.330.250.28
Denmark0.030.000.000.050.020.080.090.000.000.050.000.000.000.00
Germany0.000.000.000.060.030.000.100.000.000.000.050.000.000.00
Estonia0.280.000.000.000.210.000.000.100.000.400.480.390.300.29
Ireland0.070.000.000.000.000.000.150.140.000.000.000.000.000.00
Greece0.000.000.000.000.000.010.000.080.050.050.020.000.050.00
Spain0.020.000.000.000.000.000.000.050.000.000.000.000.000.00
France0.010.000.000.000.000.000.020.030.000.000.000.000.000.00
Croatia0.170.120.200.170.150.240.000.000.000.000.000.000.000.00
Italy0.000.000.000.000.000.000.040.000.000.000.000.000.000.00
Cyprus0.000.150.230.260.000.000.000.000.000.000.000.000.000.00
Latvia0.630.000.000.000.360.280.380.360.280.340.440.350.300.22
Lithuania0.490.520.520.530.550.400.330.380.210.320.460.390.270.26
Luxembourg0.000.000.000.000.120.000.140.000.000.000.000.000.100.00
Hungary0.420.390.340.320.360.300.310.320.220.130.300.250.110.16
Malta0.000.000.000.000.000.000.000.000.000.000.000.000.000.00
Netherlands0.000.000.000.000.000.000.000.000.000.000.000.000.000.00
Austria0.030.000.000.000.010.000.050.000.000.000.020.000.000.01
Poland0.190.000.000.000.000.000.000.000.000.000.000.000.000.00
Portugal0.000.000.000.050.000.000.000.000.000.030.000.000.060.00
Romania0.000.080.000.000.010.000.200.000.000.090.000.220.000.11
Slovenia0.310.270.270.180.270.060.180.160.000.210.220.250.180.23
Slovakia0.490.510.470.450.410.290.370.360.270.170.280.320.240.26
Finland0.300.280.280.280.310.240.310.370.330.350.350.350.300.31
Sweden0.190.030.090.290.180.080.210.180.120.150.270.130.160.14
United Kingdom0.170.360.390.360.000.420.490.470.410.000.000.000.000.00
Source: Own elaboration.

References

  1. Arrow, K.; Bolin, B.; Costanza, R.; Dasgupta, P.; Folke, C.; Holling, C.S.; Jansson, B.-O.; Levin, S.; Maler, K.-G.; Perrings, C.; et al. Economic growth, carrying capacity, and the environment. Ecol. Econ. 1995, 268, 520–521. [Google Scholar] [CrossRef]
  2. Han, X.; Chatterjee, L. Impacts of Growth and Structural Change on CO2 Emissions of Developing Countries. World Dev. 1997, 25, 395–407. [Google Scholar] [CrossRef]
  3. Beckerman, W. Economic growth and the environment: Whose growth? whose environment? World Dev. 1992, 20, 481–496. [Google Scholar] [CrossRef]
  4. Handl, G. Declaration of the United Nations Conference on the Human Environment (Stockholm Declaration), 1972 and the Rio Declaration on Environment and Development, 1992; United Nations Audiovisual Library International Law; Codification Division, Office of Legal Affairs: New York, NY, USA, 2012; pp. 1–11. (In English) [Google Scholar]
  5. Stern, D.I. Environmental Kuznets curve. Encycl. Energy 2004, 2, 1–3. [Google Scholar]
  6. Stern, D.I. The Rise and Fall of the Environmental Kuznets Curve. World Dev. 2004, 32, 1419–1439. [Google Scholar] [CrossRef]
  7. Kaika, D.; Zervas, E. The Environmental Kuznets Curve (EKC) theory-Part A: Concept, causes and the CO2 emissions case. Energy Policy 2013, 62, 1392–1402. [Google Scholar] [CrossRef]
  8. Stern, D.I. The environmental Kuznets curve after 25 years. J. Bioecon. 2017, 19, 7–28. [Google Scholar] [CrossRef]
  9. Pretty, J.; Brett, C.; Gee, D.; Hine, R.; Mason, C.; Morison, J.; Rayment, M.; Van Der Bijl, G.; Dobbs, T. Policy Challenges and Priorities for Internalizing the Externalities of Modern Agriculture. J. Environ. Plan. Manag. 2001, 44, 263–283. [Google Scholar] [CrossRef]
  10. Ceddia, M.G.; Sedlacek, S.; Bardsley, N.O.; Gomez-y-Paloma, S. Sustainable agricultural intensification or Jevons paradox? The role of public governance in tropical South America. Glob. Environ. Chang. 2013, 23, 1052–1063. [Google Scholar] [CrossRef]
  11. Culas, R.J. Deforestation and the environmental Kuznets curve: An institutional perspective. Ecol. Econ. 2007, 61, 429–437. [Google Scholar] [CrossRef]
  12. Managi, S. Are there increasing returns to pollution abatement? Empirical analytics of the Environmental Kuznets Curve in pesticides. Ecol. Econ. 2006, 58, 617–636. [Google Scholar] [CrossRef]
  13. Edenhofer, O.; Pichs-Madruga, R.; Sokona, Y.; Kadner, S.; Minx, J.; Brunner, S.; Agrawala, S.; Baiocchi, G.; Bashmakov, I.A.; Blanco, G.; et al. Climate Change 2014: Mitigation of Climate Change; Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change; Intergovernmental Panel on Climate Change: Geneva, Switzerland, 2014. [Google Scholar]
  14. Eurostat Agriculture—Greenhouse Gas Emission Statistics. Available online: http://ec.europa.eu/eurostat/statistics-explained/index.php/Agriculture_-_greenhouse_gas_emission_statistics (accessed on 29 September 2017).
  15. Fort, R.; Henrar, M. Agriculture, Forestry and Fishery Statistics, 2016th ed.; Eurostat: Luxembourg, 2016; ISBN 978-92-79-63350-8. [Google Scholar]
  16. Yan, Q.; Yin, J.; Baležentis, T.; Makutėnienė, D.; Štreimikienė, D. Energy-related GHG emission in agriculture of the European countries: An application of the Generalized Divisia Index. J. Clean. Prod. 2017, 164, 686–694. [Google Scholar] [CrossRef]
  17. Communication from the Commission to the European Council and the European Parliament: An Energy Policy for Europe 2007. Avaliable online: http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=LEGISSUM:l27067 (accessed on 27 November 2017).
  18. Christensen, L.A. Soil, Nutrient, and Water Management Systems Used in U.S. Corn Production; Agriculture Information Bulletin No. 774; United States Department of Agriculture: Washington, DC, USA, 2002.
  19. Goodland, R.; Daly, H. Environmental Sustainability: Universal and Non-Negotiable. Ecol. Appl. 1996, 6, 1002–1017. [Google Scholar] [CrossRef]
  20. Färe, R.; Grosskopf, S.; Tyteca, D. An activity analysis model of the environmental performance of firms—Application to fossil-fuel-fired electric utilities. Ecol. Econ. 1996, 18, 161–175. [Google Scholar] [CrossRef]
  21. Suri, V.; Chapman, D. Economic growth, trade and energy: Implications for the environmental Kuznets curve. Ecol. Econ. 1998, 25, 195–208. [Google Scholar] [CrossRef]
  22. Tamazian, A.; Chousa, J.P.; Vadlamannati, K.C. Does higher economic and financial development lead to environmental degradation: Evidence from BRIC countries. Energy Policy 2009, 37, 246–253. [Google Scholar] [CrossRef]
  23. Bohringer, C.; Jochem, P.E.P. Measuring the immeasurable—A survey of sustainability indices. Ecol. Econ. 2007, 63, 1–8. [Google Scholar] [CrossRef]
  24. Picazo-Tadeo, A.J.; Beltrán-Esteve, M.; Gómez-Limón, J.A. Assessing eco-efficiency with directional distance functions. Eur. J. Oper. Res. 2012, 220, 798–809. [Google Scholar] [CrossRef]
  25. Oude Lansink, A.; Wall, A. Frontier models for evaluating environmental efficiency: An overview. Econ. Bus. Lett. 2014, 3, 43–50. [Google Scholar] [CrossRef]
  26. De Koeijer, T.J.; Wossink, G.A.A.; Struik, P.C.; Renkema, J.A. Measuring agricultural sustainability in terms of efficiency: The case of Dutch sugar beet growers. J. Environ. Manag. 2002, 66, 9–17. [Google Scholar] [CrossRef]
  27. Farrell, M.J. The Measurement of Productive Efficiency. J. R. Stat. Soc. Ser. A 1957, 120, 253–290. [Google Scholar] [CrossRef]
  28. Lauwers, L. Justifying the incorporation of the materials balance principle into frontier-based eco-efficiency models. Ecol. Econ. 2009, 68, 1605–1614. [Google Scholar] [CrossRef]
  29. Vlontzos, G.; Niavis, S. Assessing the Evolution of Technical Efficiency of Agriculture in EU Countries: Is There a Role for the Agenda 2000? In Agricultural Cooperative Management and Policy: New Robust, Reliable and Coherent Modelling Tools; Zopounidis, C., Kalogeras, N., Mattas, K., van Dijk, G., Baourakis, G., Eds.; Springer International Publishing: Cham, Switzerland, 2014; pp. 339–351. ISBN 978-3-319-06635-6. [Google Scholar]
  30. Vlontzos, G.; Niavis, S.; Manos, B. A DEA approach for estimating the agricultural energy and environmental efficiency of EU countries. Renew. Sustain. Energy Rev. 2014, 40, 91–96. [Google Scholar] [CrossRef]
  31. Armundito, E.; Kaneko, S. Environmental Productivities and Carbon Abatement Costs of Manufacturing Sectors. In Climate Change Policies and Challenges in Indonesia; Kaneko, S., Kawanishi, M., Eds.; Springer: Tokyo, Japan, 2016; Volume 9, pp. 199–218. [Google Scholar]
  32. Scheel, H. Undesirable outputs in efficiency valuations. Eur. J. Oper. Res. 2001, 132, 400–410. [Google Scholar] [CrossRef]
  33. Rodseth, K.L. Environmental efficiency measurement and the materials balance condition reconsidered. Eur. J. Oper. Res. 2016, 250, 342–346. [Google Scholar] [CrossRef]
  34. Knox Lovell, C.A.; Pastor, J.T.; Turner, J.A. Measuring macroeconomic performance in the OECD: A comparison of European and non-European countries. Eur. J. Oper. Res. 1995, 87, 507–518. [Google Scholar] [CrossRef]
  35. Vlontzos, G.; Pardalos, P.M. Assess and prognosticate green house gas emissions from agricultural production of EU countries, by implementing, DEA Window analysis and artificial neural networks. Renew. Sustain. Energy Rev. 2017, 76, 155–162. [Google Scholar] [CrossRef]
  36. Halkos, G.E.; Tzeremes, N.G.; Kourtzidis, S.A. Regional sustainability efficiency index in Europe: An additive two-stage DEA approach. Oper. Res. 2015, 15, 1–23. [Google Scholar] [CrossRef]
  37. Chung, Y.H.; Färe, R.; Grosskopf, S. Productivity and Undesirable Outputs: A Directional Distance Function Approach. J. Environ. Manag. 1997, 51, 229–240. [Google Scholar] [CrossRef]
  38. Watanabe, M.; Tanaka, K. Efficiency analysis of Chinese industry: A directional distance function approach. Energy Policy 2007, 35, 6323–6331. [Google Scholar] [CrossRef]
  39. Suzuki, S.; Nijkamp, P. An evaluation of energy-environment-economic efficiency for EU, APEC and ASEAN countries: Design of a Target-Oriented DFM model with fixed factors in Data Envelopment Analysis. Energy Policy 2016, 88, 100–112. [Google Scholar] [CrossRef]
  40. Tyteca, D. Linear programming models for the measurement of environmental performance of firms—Concepts and empirical results. J. Product. Anal. 1997, 8, 183–197. [Google Scholar] [CrossRef]
  41. Zaim, O. Measuring environmental performance of state manufacturing through changes in pollution intensities: A DEA framework. Ecol. Econ. 2004, 48, 37–47. [Google Scholar] [CrossRef] [Green Version]
  42. Zhou, P.; Ang, B.W.; Wang, H. Energy and CO2 emission performance in electricity generation: A non-radial directional distance function approach. Eur. J. Oper. Res. 2012, 221, 625–635. [Google Scholar] [CrossRef]
  43. Sueyoshi, T.; Goto, M. Data envelopment analysis for environmental assessment: Comparison between public and private ownership in petroleum industry. Eur. J. Oper. Res. 2012, 216, 668–678. [Google Scholar] [CrossRef]
  44. Zhou, P.; Ang, B.W.; Poh, K.L. A survey of data envelopment analysis in energy and environmental studies. Eur. J. Oper. Res. 2008, 189, 1–18. [Google Scholar] [CrossRef]
  45. Sueyoshi, T.; Yuan, Y.; Goto, M. A literature study for DEA applied to energy and environment. Energy Econ. 2017, 62, 104–124. [Google Scholar] [CrossRef]
  46. Coelli, T.J.; Rao, D.S. Total factor productivity growth in agriculture: A Malmquist index analysis of 93 countries, 1980–2000. Agric. Econ. 2005, 32, 115–134. [Google Scholar] [CrossRef]
  47. Zhu, X.; Lansink, A.O. Impact of CAP subsidies on technical efficiency of crop farms in Germany, the Netherlands and Sweden. J. Agric. Econ. 2010, 61, 545–564. [Google Scholar] [CrossRef]
  48. Martinho, V.J.P.D. Efficiency, total factor productivity and returns to scale in a sustainable perspective: An analysis in the European Union at farm and regional level. Land Use Policy 2017, 68, 232–245. [Google Scholar] [CrossRef]
  49. Toma, P.; Miglietta, P.P.; Zurlini, G.; Valente, D.; Petrosillo, I. A non-parametric bootstrap-data envelopment analysis approach for environmental policy planning and management of agricultural efficiency in EU countries. Ecol. Indic. 2017, 83, 132–143. [Google Scholar] [CrossRef]
  50. Lauwers, L.; Van Huylenbroeck, G.; Rogiers, G. Technical, economic and environmental efficiency analysis of pig fattening farms. In Proceedings of the 9th European Congress of Agricultural Economists, Warschau, Polen, 24–28 August 1999. [Google Scholar]
  51. Hoang, V.N.; Rao, D.S.P. Measuring and decomposing sustainable efficiency in agricultural production: A cumulative exergy balance approach. Ecol. Econ. 2010, 69, 1765–1776. [Google Scholar] [CrossRef]
  52. Tian, D.; Zhao, F.; Mu, W.; Kanianska, R.; Feng, J. Environmental efficiency of chinese open-field grape production: An evaluation using data envelopment analysis and spatial autocorrelation. Sustainability 2016, 8, 1246. [Google Scholar] [CrossRef]
  53. Ghali, M.; Latruffe, L.; Daniel, K. Efficient use of energy resources on French farms: An analysis through technical efficiency. Energies 2016, 9, 601. [Google Scholar] [CrossRef]
  54. Halkos, G.E.; Tzeremes, N.G. Exploring the existence of Kuznets curve in countries’ environmental efficiency using DEA window analysis. Ecol. Econ. 2009, 68, 2168–2176. [Google Scholar] [CrossRef] [Green Version]
  55. Zaim, O.; Taskin, F. A Kuznets curve in environmental efficiency: An application on OECD countries. Environ. Resour. Econ. 2000, 17, 21–36. [Google Scholar] [CrossRef]
  56. Oggioni, G.; Riccardi, R.; Toninelli, R. Eco-efficiency of the world cement industry: A data envelopment analysis. Energy Policy 2011, 39, 2842–2854. [Google Scholar] [CrossRef] [Green Version]
  57. Galeotti, M.; Lanza, A.; Pauli, F. Reassessing the environmental Kuznets curve for CO2 emissions: A robustness exercise. Ecol. Econ. 2006, 57, 152–163. [Google Scholar] [CrossRef]
  58. Hoechle, D. Robust standard errors for panel regressions with cross-sectional dependence. Stata J. 2010, 10, 288–308. [Google Scholar]
  59. Eurostat. Available online: http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=apro_acs_a&lang=en (accessed on 1 September 2017).
  60. Economic Accounts for Agriculture—Values at Current Prices. Available online: http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=aact_eaa01&lang=en (accessed on 1 September 2017).
  61. Eurostat Database. Available online: http://ec.europa.eu/eurostat/data/database (accessed on 1 September 2017).
  62. Haug, R. Some leading issues in agricultural extension: A literature review. J. Agric. Educ. Ext. 1999, 5, 263–274. [Google Scholar] [CrossRef]
Figure 1. Average Eco-(in)efficiency of EU agricultural sector and average Eco-(in)efficiency by groups of countries according to their date of accession (1999–2012) (Source: Own elaboration).
Figure 1. Average Eco-(in)efficiency of EU agricultural sector and average Eco-(in)efficiency by groups of countries according to their date of accession (1999–2012) (Source: Own elaboration).
Energies 10 01992 g001
Figure 2. The Turning Points of Models 2.1, 2.2. and 2.3 (a) The turning points of Eco-(in)efficiency curve (b) The turning points of the energy use curve (c) The turning point of the GHG curve (Source: Own elaboration).
Figure 2. The Turning Points of Models 2.1, 2.2. and 2.3 (a) The turning points of Eco-(in)efficiency curve (b) The turning points of the energy use curve (c) The turning point of the GHG curve (Source: Own elaboration).
Energies 10 01992 g002
Table 1. Descriptive statistics of DEA inputs and outputs.
Table 1. Descriptive statistics of DEA inputs and outputs.
StatisticLand (1000 Ha)Energy (mil. $)Chemicals and Fertilizers (mil. $)Fixed Capital Consumption (mil. $)Labour (1000 AWU)Output (mil. $)GHG Emissions (1000 tonnesCO2)
Mean6642.64786.57880.121815.241210.8811,720.4517.69
St Dv7901.59933.031314.672730.811729.0615,644.3722.99
Max35,177.804502.707599.1012,377.397307.3570,394.90100.46
Min9.705.441.733.763.59115.180.08
Source: [59,60,61]; Own elaboration.
Table 2. Eco-(in)efficiency scores and longitudinal performance in EU agricultural sector (1999–2012).
Table 2. Eco-(in)efficiency scores and longitudinal performance in EU agricultural sector (1999–2012).
CountryAverage Eco-(In)EfficiencyYears Fully EfficientHighest Eco-(In)EfficiencyLowest Eco-(In)Efficiency
Netherlands0.000140.0000.000
Malta0.000140.0000.000
Italy0.003130.0430.000
France0.004110.0300.000
Spain0.005120.0500.000
Austria0.00990.0550.000
Belgium0.010100.0680.000
Portugal0.010100.0580.000
Poland0.013130.1880.000
Germany0.017100.0980.000
Greece0.01870.0770.000
Bulgaria0.021100.1330.000
Denmark0.02280.0880.000
Luxembourg0.025110.1360.000
Ireland0.026100.1530.000
Cyprus0.046110.2550.000
Romania0.05080.2160.000
Croatia0.07580.2410.000
Sweden0.15900.2860.029
Estonia0.17560.4790.000
Slovenia0.19910.3120.000
Czech Republic0.21950.4510.000
United Kingdom0.22060.4920.000
Hungary0.28100.4220.113
Latvia0.28230.6340.000
Finland0.31100.3750.239
Slovakia0.35100.5130.167
Lithuania0.40200.5510.213
Mean0.105---
St.Dv0.150---
Source: Own elaboration.
Table 3. Models 2.1, 2.2 and 2.3 estimation results.
Table 3. Models 2.1, 2.2 and 2.3 estimation results.
Models (Dependent Variable)
2.1 ( e c o E f f )2.2 ( E N E R p c )2.3 ( G h G p c )
EstimatorsEstimationSt.ErrorEstimationSt.ErrorEstimationSt.Error
β G D P 0.0428 ***0.01451.759 **0.71690.0060 ***0.00195
β G D P 2 −2.62 × 10−9 ***7.75 × 10−10−1.06 × 107 ***3.92 × 10−8−2.43 × 10−10 **1.09 × 10−10
β G D P 3 4.30 × 10−14 ***1.27 × 10−141.89 × 10−12 ***6.55 × 10−131.65 × 10−151.82 × 10−15
a −0.03990.0820−2.63294.045081.5590 ***15.490
T14
N27
WaldTest chi2(1)8.689.75137.46
Prob > chi20.0030.0210.001
Hausman Test chi2(1)-0.013.84
Prob > chi2-0.9430.051
TurningPoint 1 (000€)11,32612,42812,325
TurningPoint 2 (000€)29,29424,961-
Statistical significance: (***) at 0.01 level (**) at 0.05 level. Source: Own elaboration.

Share and Cite

MDPI and ACS Style

Vlontzos, G.; Niavis, S.; Pardalos, P. Testing for Environmental Kuznets Curve in the EU Agricultural Sector through an Eco-(in)Efficiency Index. Energies 2017, 10, 1992. https://doi.org/10.3390/en10121992

AMA Style

Vlontzos G, Niavis S, Pardalos P. Testing for Environmental Kuznets Curve in the EU Agricultural Sector through an Eco-(in)Efficiency Index. Energies. 2017; 10(12):1992. https://doi.org/10.3390/en10121992

Chicago/Turabian Style

Vlontzos, George, Spyros Niavis, and Panos Pardalos. 2017. "Testing for Environmental Kuznets Curve in the EU Agricultural Sector through an Eco-(in)Efficiency Index" Energies 10, no. 12: 1992. https://doi.org/10.3390/en10121992

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop