1. Introduction
Assessing and improving the sustainability of farming is an issue of growing importance, especially because farms, and, more precisely, the cradle-to-farm-gate link of the food chain, play a major role in the environmental impact generation of the entire chain (see e.g., [
1,
2,
3,
4,
5,
6]). Complying with the carrying capacity of both local and global ecosystems is a prerequisite to ensure sustainability [
7]. Based on theoretical considerations and using the local versus global carrying capacity distinction as a starting point, Repar et al. [
7] developed a framework for the assessment of environmental performance at farm level and thereby distinguished between the local and global environmental performance of a farm. Farm global environmental performance is defined as the environmental intensity of agricultural production in the cradle-to-farm-gate link of the food chain [
7]. Environmental intensity is measured as the global (i.e., on- and off-farm) environmental impact generation per unit of biophysical farm output (e.g., digestible energy produced for humans by the farm) [
7]. Local environmental performance is measured by the on-farm environmental impact generation per unit of farm usable agricultural area [
7].
The distinction between global and local environmental performance proposed by Repar et al. [
7] builds upon previous considerations made by several authors in this field [
8,
9,
10,
11,
12,
13]. All these previous contributions acknowledged the need to distinguish between two major types of environmental issues (local/regional versus global) depending on the scale of environmental relevance of the impacts associated with each issue. They also advocated the use of different types of environmental performance indicators depending on that scale. Both local and global environmental scales should be considered simultaneously to avoid problem shifting from one scale to another [
7,
12,
13]. The approach proposed by Repar et al. [
7] for farm environmental performance assessment further developed the existing considerations available in the literature on this topic. It also embedded them in a theoretical framework relying on the ecosystem’s carrying capacity concept, which is a central pillar of the environmental sustainability concept.
Better understanding of the relationship between local and global environmental performance and between environmental and economic performance at farm level is highly relevant for improving the sustainability of farming. This is particularly important from the agricultural policy perspective. The promotion of sustainable agriculture requires implementation of appropriate policy instruments that enhance both local and global farm environmental performance. However, up to now, farm-level agricultural policy instruments have mostly focused on screening and improvement of what could be referred to as local environmental performance, e.g., nitrogen surplus per ha (see for instance [
14,
15,
16,
17,
18,
19]). The relationship between the local and global dimension of farm environmental performance has not been investigated in the literature and is therefore unknown. Consequently, we have no guarantee that these agri-environmental policy measures intended to improve the local environmental performance of farms also lead to an improved global environmental performance.
Simultaneously with the improvements in farm environmental performance, achieving agricultural sustainability also requires improvements in the economic performance of farming. The relationship between farm environmental and economic performance has already been investigated in a few studies relying on life cycle assessment [
11,
20,
21,
22]. With the exception of Jan et al. [
11] who explicitly focused on farm global environmental performance as specified in Repar et al. [
7], none of these contributions explicitly differentiated between the local and global environmental performance of a farm. However, given the type of environmental performance indicators used, these three contributions implicitly all addressed—to a more or less narrow extent—the global environmental performance of a farm as defined in Repar et al. [
7] and its relationship to farm economic performance. Furthermore, also with the exception of Jan et al. [
11], none of these studies used complete economic performance indicators that would consider all production factors. Despite differences regarding the economic performance indicators used and the types of farms investigated, these four investigations all found a positive relationship between global environmental and economic performance of farming. Thus they all highlighted the existence of a synergy between these two dimensions of sustainable performance of a farm. However, as is obvious from this overview, a study of the relationship between local environmental performance and economic performance of a farm is lacking.
The objective of our research, which assessed Swiss dairy farms in the alpine area building upon the work of Jan et al. [
11], was twofold. Firstly, it aimed to investigate the relationship between the local and global environmental performance of these farms and to highlight possible synergies or trade-offs in the promotion of these two dimensions of farm environmental performance. The second objective was to comprehensively analyse the link between the environmental and economic performance of these farms. We divided this second objective into two sub-objectives. The first one consisted of broadening the analysis carried out by Jan et al. [
11] on the relationship between farm global environmental and economic performance. The second sub-objective was to examine the link between the local environmental and economic performance of the sample farms.
4. Discussion
This section discusses the main findings of our investigation, firstly summarizing them and then relating them to other studies in the field. We finish this section by addressing the limitations of our work.
4.1. Main Findings
In the present work, we applied—within a case study for Swiss alpine dairy farms—the approach proposed by Repar et al. [
7] to assess farm environmental performance with a differentiation between farm local and global environmental performance. To assess the local environmental performance of a farm, we decomposed the cradle-to-farm-gate impacts assessed by means of LCAs into their on- and off-farm parts. We considered a very broad set of environmental impact categories to provide the fullest possible environmental performance picture.
The analysis of the link between farm local and global environmental performance revealed complex relationships. Depending on the environmental impact categories considered, no significant relationships, trade-offs and synergies were observed. However, trade-offs were more frequent than synergies. Furthermore, we found synergies between farm global environmental performance and farm economic performance, regardless of the environmental impact category observed or the indicator of economic performance chosen. For most impact categories considered, the analysis showed no significant relationship between local environmental performance and economic performance, with very few exceptions, where a weak synergy or trade-off existed.
4.2. Discussion of the Main Findings
This work represents the first implementation of the framework proposed by Repar et al. [
7] to assess environmental performance at farm level. To the best of our knowledge, this is therefore the first study that distinguishes between the local and global dimensions of the environmental performance of a farm and comprehensively analyses their mutual link as well as their relationship with farm economic performance. Very few studies analysing the relationship between economic and environmental performance at farm level can be found in the literature (see, for instance, [
11,
21]). However, these studies focused solely on what Repar et al. [
7] called “farm global environmental performance” and analysed its link to farm economic performance.
Our finding of a positive relationship between farm global environmental and economic performance is similar to that of Jan et al. [
11], who also found a synergy between these two dimensions of the sustainable performance of a farm. Jan et al. [
11] used the same original dataset but relied (i) on an older SALCA version for the environmental impact assessment and (ii) on only one economic performance indicator, namely work income per family work unit. As discussed in Jan et al. [
11], three other contributions [
20,
21,
50] also reported a positive relationship between global environmental performance and economic performance.
No study explicitly analysed the link between farm local environmental and economic performance, which likely has two major reasons. First, the distinction between local and global environmental performance has only recently been introduced [
7]. Second, almost all empirical LCA applications have up to now—due to the life cycle perspective inherent to LCA—exclusively dealt with global environmental impacts as defined in
Section 2.4. Nevertheless, some results regarding this link can be found in Thomassen et al. [
21] for Dutch dairy farms. However, this link was not the focus of their investigation, and the two indicators we can identify as “local environmental performance indicators” were not even referred to as such in the publication. Moreover, Thomassen et al. [
21] used partial economic performance indicators that did not consider all production factors and therefore did not reflect the overall economic performance of a farm. Despite this methodological difference to our study, it is interesting that Thomassen et al. [
21] found no correlation between the farm local environmental performance indicators (on-farm eutrophication per ha, on-farm acidification per ha) and the economic performance indicators (gross value added per kg milk, gross value added per unit of labour). Their finding is therefore similar to ours that also reveals mostly no significant correlation between the local environmental performance indicators and three different (complete) indicators of economic performance. Our findings regarding the relationship between farm global and local environmental performance cannot be compared with those of similar studies because such studies do not exist for the reasons mentioned at the beginning of this section.
4.3. Implications of Our Findings for the Sustainable Intensification Debate
There exists an extensive body of scientific literature dedicated to the comparison of environmental impacts of intensive and extensive agricultural systems (see e.g., [
51,
52,
53]). In the last decade, the sustainable intensification concept came to the forefront of the debate on the future of agriculture. This debate is especially focused on the degree of agricultural intensity and the future challenge of feeding a growing and increasingly wealthy human population. The sustainable intensification concept actually “originates from sub-Saharan agriculture in the 1990s and originally focussed on building adaptable farming systems that support the livelihood of the rural poor” [
54]. In the last decade, its meaning has shifted towards the “enhancement of agricultural productivity while reducing environmental impacts” or, in more operationalised terms, “the production of more food with less resources” (adapted from Rockström et al. [
55]). The sustainable intensification discussion has thus mostly targeted improvements in agricultural sustainability at the global level (see e.g., [
55,
56,
57,
58]), or what we call the global environmental performance (or eco-efficiency) of farming.
Our work does not primarily aim at comparing the environmental performance of dairy farming systems with different production intensities. However, it indirectly has substantial implications for the debate on the sustainable intensification of farming. The local environmental performance defined and assessed in our work is strongly connected to the farming intensity because it measures the extent of the local environmental impact generation per hectare usable agricultural area. It is thus an indicator of the “local environmental burdens” resulting from the farming intensity. Our findings of the existence of negative correlations between local and global environmental performance imply, at least for the Swiss dairy farms of the mountain area, that an improvement of the global environmental performance will likely lead to a deterioration of the local environmental performance. This means that the sustainable intensification debate, due to its unilateral focus on global environmental performance, will most likely not lead to a holistic environmental sustainability improvement in agriculture but to food chains that are globally more eco-efficient but locally worse off in environmental terms.
We therefore advocate the following redefinition of sustainable intensification: “Sustainable intensification aims at improving the biophysical eco-efficiency of food production over the whole food chain (global environmental performance) while at the same time ensuring that the environmental impacts generated at the local level do not exceed the carrying capacities of the local ecosystems (local environmental performance)”. Due to the existence of the aforementioned trade-offs between the global and local dimension of farm environmental performance, the challenge for sustainable intensification is to find technologies that enable simultaneous improvement in both dimensions.
4.4. Limitations and Future Research Need
Although the framework established by Repar et al. [
7] and used here can be implemented to various farms, irrespective of their type or location, it is important to emphasise that the findings of the present empirical study apply only to Swiss dairy farming in the alpine area. Furthermore, as pointed out by Jan et al. [
11], because the sample used for this study was quite small and not selected at random, there are limitations regarding its representativeness [
11]. These limitations should be considered when interpreting the results of this work.
As discussed in Repar et al. [
7], some issues of conceptual nature also arise when using the framework for farm environmental performance assessment implemented in the present paper. Firstly, because we focused on the cradle-to-farm-gate analysis, the subsequent parts of the chain, which are important for painting the wholesome sustainability picture, were ignored. Although focusing on the production perspective provides an important view, there are also other strategies to improve the sustainability of the food chain that have to be considered on the consumption side. The examples of such strategies are the reduction of food waste (see, for example, Gentil et al. [
59]) or the change in diets (see, for example, Tukker et al. [
60]).
Secondly, as mentioned in the introduction, in our framework, the local versus global carrying capacity distinction was used only as a starting point for the differentiation between local and global environmental performance. However, the indicators proposed did not directly integrate the carrying capacity constraints and are therefore of a relative nature [
7]. Such indicators enable a relative improvement in terms of sustainability but are still no guarantee for the achievement of an absolute sustainable state [
7]. As identified by Sala et al. [
61], further research and development of the methods in the LCA field are needed in order for the indicators to better reflect the carrying capacity and planetary boundaries. First LCA-based approaches that integrate carrying capacities into environmental performance indicators and enable analyses to move from relative to absolute environmental sustainability were recently developed (e.g., [
62,
63,
64]). Also, Repar et al. [
7] developed conceptual considerations on how to integrate carrying capacities into the indicators of the local and global environmental performance they proposed. The practical implementation of these carrying capacities should be the subject of future research work.
Finally, our work did not account for the third dimension of sustainability, namely the social one. Future work should therefore assess the link between (i) local environmental performance and social performance; (ii) global environmental performance and social performance; and (iii) economic performance and social performance, in order to provide a complete sustainability overview. Such assessment requires the implementation of the social sustainability concept into farm-level indicators of social performance. This implementation is probably as challenging as the development of theoretically sound farm-level environmental performance indicators.
Analysing the relationship between different performance dimensions is a first important contribution to a deeper understanding of farm sustainability. However, what is ultimately necessary for practical improvements is to understand the mechanisms behind these relationships. Furthermore, farm management strategies and production technologies that enable simultaneous improvements in global and local environmental and economic performance of farming need to be identified. This calls for very detailed investigations of the factors affecting farm environmental and economic performance.
5. Conclusions
Our analysis provides evidence that the improvement of the environmental sustainability of dairy farming in the mountain region of Switzerland is a highly complex endeavour. Both synergies and trade-offs exist between the local and global environmental performance of a farm, depending on the environmental issue considered. Interestingly, the often raised and feared possible trade-off between environmental and economic performance could not be confirmed empirically, neither for the local nor for the global dimension of environmental performance. Contrariwise, we found synergies between farm global environmental and economic performance. This implies that the improvement of the eco-efficiency of food production in the cradle-to-farm-gate link of the food chain is very likely to lead to an improvement of the economic performance and vice versa.
The complex relationships between farm local and global environmental performance imply that no one-size-fits-all solution may exist for the improvement of farm environmental sustainability. The results suggest that exclusively focusing on the global environmental performance, i.e., on the eco-efficiency of food production in the cradle-to-farm-gate link of the food chain could negatively affect the local environmental performance. To avoid that any improvement in one dimension of environmental performance happens at the expense of the other, both local and global performance dimensions have to be considered. Life cycle assessment (LCA) practitioners should therefore be aware of the potential prejudicial side effects of a unilateral focus on global environmental performance. A holistic farm environmental performance assessment encompassing both local and global environmental performance dimensions calls for a standard decomposition into on- and off-farm impacts in LCIA tools.
Furthermore, our findings have implications for policy makers. Existing farm-level agri-environmental policy measures and instruments in Switzerland, as in many other countries, tend to focus exclusively on the local dimension of farm environmental performance. Due to the negative correlations that were found between local and global environmental performance, these instruments may lead to a deterioration of farm global environmental performance. Hence, a clear definition of the objectives of environmental policy measures, the consideration of both local and global aspects of environmental performance and the use of LCAs in policy making are indispensable. These actions are required if we wish to prevent problem shifting between the local and global ecosystems and reach real improvements in terms of environmental sustainability. The necessity of considering the two dimensions of environmental performance also applies for the development and assessment of new agricultural technologies intended to improve the environmental sustainability of farming.
Finally, from a more general perspective, our findings have potentially far-reaching implications, especially if these findings should be confirmed for other types of farms and countries. As mentioned previously, when dealing with the environmental sustainability of farming, scientists and policy makers have until now been adopting a one-sided focus on either global environmental performance (for instance, LCA practitioners) or on local environmental performance (for instance, most farm-level agri-environmental policy makers). Through this one-sided focus, they implicitly assumed that local and global environmental performance go hand in hand and do not need to be considered separately. Our finding of the existence of trade-offs between farm local and global environmental performance refutes—at least for Swiss dairy farming—this widespread assumption. In that sense, our work indirectly questions whether these one-sided perspectives, which have been used widely for years, have always been able to reach real improvements in terms of environmental sustainability.