Influence of decoupling of CAP subsidies on
agricultural production within the EU
Author: Katarina Cizmarikova Student number: 1761579
Supervisor: Dr. P.W. van Wijck
Master thesis
Leiden University, 2017-06-08
Public Administration Economics and governance stream
2
Table of Content
Executive summary ... 3 1. Introduction ... 4 1.1. Policy background ... 5 1.1. Research objective ... 8 1.2. Hypothesis ... 9 2. Policy foundations ... 11 2.1. Policy evolution ... 11 3. Theoretical framework ... 14 3.1. Impact of decoupling ... 14 4. Research design ... 21 4.1. Operationalization ... 224.2. Data and limitations ... 24
4.3. Production trends ... 26 5. Analysis ... 36 5.1. Results ... 36 5.2. Discussion ... 45 6. Conclusion ... 48 7. References ... 50 8. Appendix ... 55
3
Executive summary
The mail aim of this thesis is to establish an impact of subsidies of the Common Agricultural Policy (CAP) on the agricultural production within the EU-15 states. This policy one of the most important ones in the EU, installed to provide food security and safety for European citizens and production support and stability for European farmers.
Since its establishment, the CAP has been experiencing regular reforms. The most radical reform, from 2003, introduced changes in a financial support for the European farmers. These changes represented decoupling of support payments from the production. In practise this means that subsidies’ recipients receive benefits regardless of their production output. The main purpose of decoupling was to reduce its intervention with trade and production decisions of farmers.
Given the importance of this measure for European agriculture and its continuous improvements, the main goal of this research is to examine the actual impact of subsidies on the production after the decoupling measure took place. The main hypothesized assumption is that decoupling has a positive effect on the production within the EU-15 member states.
Initially, research examined the policy’s foundations that helped to comprehend the main mechanics and elements that shaped the decoupling measure. Subsequently, an existent theoretical framework focusing on an investigation of decoupling effects was reviewed. Results from this framework, which was structured according to the central hypothesis, revealed that decoupling measure has on the production predominantly positive effects. In research design, a panel data regression analysis was selected as the main method of research, with Eurostat database serving as the main source of data.
The results from the hypothesis-testing analysis have corresponded to the central assumption. They also confirmed the claims of the reviewed theory, that decoupling has positive effects on the aggregated total production outputs after the change of the policy, of both plant commodities and livestock goods within the EU-15 member states.
4
1. Introduction
This thesis examines the impact of subsidies of the Common Agricultural Policy (CAP) on the agricultural production within the EU-15 states.
The effects of increasing price volatility, demographic growth, as well as changes in lifestyles, diets and incomes cause a steady increase in a global demand for food. According to the Technical Centre for Agricultural and Rural Cooperation (CTA), in 2016 were 240 million people across 45 countries experiencing food stress, of which 80 million were classified as being in a food crisis (CTA, 2016). Various external factors continuously impact food security such as global conflicts, migration, socio-economic fluctuations or climatic developments. In order to combat these elements an increased agricultural production is needed that would tackle food crises and support affected countries, both developed and developing.
The CAP is the European Union (EU)’s system of agricultural subsidies and programmes that is installed to provide food security through a stable agricultural production that responds to the market demand. The policy is overlooking the EU’s agricultural markets and covers areas of farming, environmental and rural developments. Main goals of the CAP include a provision of a safe food supply and security, support of European farmers to produce in sustainable and stable manner, as well as to protect environmental and animal welfare (EC, 2016). Even though the budget share on the policy has decreased over the past 25 years, from 73% in 1985 to 40% in 2016, it still represents a major expense within the EU’s budget (EC, 2017).
Since its inception, the CAP has been experiencing regular modifications and amendments. The most radical reform that involved changes in financial support of the EU farmers occurred during the CAP 2003 reform. It entailed decoupling of direct support payments from the production, meaning that recipients of subsidies receive payments regardless of the amount of their production. The main purpose of decoupling agricultural policies is to cut down their intervention with production decisions and trade (OECD, 2005). Given the importance of the policy for the European agriculture, it is interesting to examine what is the actual impact of the CAP subsidies on the production of the key agricultural commodities after the decoupling measure took place.
5 The structure of this thesis will be outlined as follows. Firstly, the background of the CAP policy will be presented in order to achieve an understanding of the main policy concepts and terms referred to throughout the thesis. After establishing the main focus of the investigation, the research objective and central hypothesis will be drawn. This is to guide the research and cover the relevant factors that affecting the phenomenon.
Secondly, the CAP foundations will be examined in greater details to understand the main mechanics and elements of the policy. This section will help to comprehend where the main research concept originated from and what shaped its current form.
Thirdly, a review of existent theoretical framework investigating effects of decoupling on production will be outlined. This will serve as a basis for the research analysis and establishment of the final results and conclusions on the influence of decoupling. It will be structured according to main hypothetical concept formulated on the beginning of the research.
Fourthly, the research design will present the type of research and data collection that will be employed to test the hypothesis. It will also provide operationalization of the concepts from the theoretical framework. Next, an overview of agricultural production trends will be reviewed to provide deeper insights into the development of the EU-15 production over the years.
Fifthly, the Analysis section will test and analyse both established hypothesis and theoretical framework. Finally, the effect of decoupling on the production will be established and final conclusions will be drawn.
1.1. Policy background
Currently of all the European institutions, agricultural sector is the only one that holds a common policy for all member states. At the present time it is responsible for over 39% of the entire EU budget (DEFRA, 2015). Privileged status that agricultural sector has was installed right after Second World War, when the main priority of European countries was provide sufficient and safe food provision for its citizens. Throughout history, the CAP has had a
6 significant role in connecting European countries, strengthening national commitments and helping to change the Europe into a major global food exporter (ERS, 2016).
However, spending of the policy’s budget has significantly changed throughout the course of its existence. Under a period of unreasonably high subsidies installed across post-war Europe, the EU agriculture flourished. This led to an overproduction of butter, grain, wine and other agricultural commodities, which created scandalous so called ,,butter mountains and wine lakes’’ owned by supposedly wealthy landowners living high lifestyles from taxpayers’ subsidies (Chrisafis, 2006; Castle, 2009; Eaton, 2012). This food surplus represented a problem for the EU policy- makers who wanted to prevent a substantial drop in prices, which would normally occur when the supply majorly escalates. The agricultural sector was heavily subsidized and thus rather expensive to maintain in long term. Therefore, as a reaction to this overproduction, the decoupling measure of production and payments was installed. Since the instalment of this measure, regular policy reforms were set to reimburse producers for negative fluctuations in price levels. This financial support is indifferent on the level of production, but it is established on the economic situation of the producer.
The policy is composed of two pillars. Measures related to market and direct payments to farmers are covered in the first pillar. Rural Development Programme that provides a framework investing in farm projects, rural activities and a promotion of competitiveness and innovation are covered in the second pillar (EC, 2016b).
Regarding the budget of the first pillar that is allocated to each member state, the overall amount for subsidies is conditioned yearly by the size of country’s financial envelope from the EU budget (EC, 2016a). After budget allocation, national governments have the possibility to establish which of the payments schemes will be financed or if a financial transfer between pillars will occur. National budget allocations of direct payments will be continuously adjusted in order to achieve more harmonized and equal distribution among the EU member states (EC, 2016a).
As the decoupling policy measure is the most important within the CAP framework, this thesis will focus on the analysis of its effect on the production within the EU-15. For the reference, the analysis will use data from the first pillar, namely the Decoupled direct aid (Article 05 03 01) that is covered in the Basic payment scheme of the CAP’s direct payments
7 subsidies (EC, 2016a). Further explanation of the decoupling concept and the CAP schemes in which it is implemented are outlined in the following sections.
Decoupling
Firstly, it is important to clearly define the concept of “decoupling”, to understand its representation in various interpretations and contexts. In general, payments that are decoupled from production represent transfers to agricultural producers, which are independent of their current production output and prices of agricultural commodities (Becvarova, 2007). This type of direct income allocation reassigns financial resources from urban to rural areas, and leads to sector changes in the economy. In policy-making terms, decoupling is understood as policy having no or only very small effects on the production or trade (Lopez, 2001). For the agricultural producers who are recipients of this support, decoupling represents an increase of income and, subsequently, of wealth. This is because subsidies provide a security of continuous and stable income revenue that is no longer dependent on their actual production output.
In 1987, the key focus of decoupling was to reduce international trade distortions while supporting the agricultural sector. Decoupling quickly became one of the main elements of agricultural policy. Later, in 1994 the Uruguay Round Agreement on Agriculture established that policies with no or small effects on production and trade are immune to any controls or regulations (Lopez, 2001). Consequently, such policies came into question and under investigations of various researchers across the globe, who tried to prove if their influence on trade and production really is neutral. Even though these studies used different theoretical approaches and compared different regimes, countries or products, they all came to a similar conclusion: decoupling of policies influence farm decisions on what and how to produce, and consequently their overall agricultural production.
In order to analyse the impact of subsidies after the decoupling, central schemes employed throughout the length of the policy have to be defined. The main schemes considered as decoupled under the CAP 2007-2013 guidelines are the single payment scheme (SPS), the single area payment scheme (SAPS) and the decoupled specific support that is under the article 68 of Council Regulation (EC) No 73/2009. The most recent regulation, Regulation
8 (EU) No 1307/2013 implemented by the CAP 2013 reform has replaced these schemes. It redefined the direct payments system as Basic Payment Scheme, effective from January 1st 2015, and repealing the two previous policy regulations, Council Regulation (EC) No 637/2008 and Council Regulation (EC) No 73/2009 (Massot, 2017).
All of these support schemes are paid out independently of the production but conditioned on the principle of cross-compliance. Cross-compliance is a set of rules on food safety, animal and plant health, environment, as well as the maintained state of a farmland (EC, 2016c). The flexibility in implementing this scheme across the EU member states means that there are different variations in their design and levels of decoupling.
1.1. Research objective
The main goal of this research is to examine if decoupling of subsidies has an influence,
specifically a positive effect, on the agricultural production within the EU-15 member countries, for main categories of particular agricultural commodities.
Therefore, the central research question is: “What is the influence of decoupling of direct
payments on the agricultural production within the EU-15 member states with regards to the output of their agricultural commodities?”
The main purpose of this thesis is to answer the abovementioned research question, and thus attempt to explain the researched phenomenon. In order to achieve this, the most relevant factors that influence different aspects of the policy support will be examined. For an establishment of the effect of subsidies on production, after the decoupling measure was implemented, data from Eurostat database will be applied. These data will help to assess production trends, amounts of allocated subsidies, and other relevant variables. The panel data will examine the period 2000–2016 and include countries of the EU-15.
The added value of the research compared to existing studies on this topic is in discovering the underlying causal mechanisms, which shape the effects of subsidies on the agricultural production within the EU-15. Existent studies have investigated the influence of decoupling but mainly in theoretical terms and empirical testing thus remained on the very limited level.
9 Most of these also focused on the period 2003-2013, and therefore did not provide information on the most recent years. This research is combining theoretical and empirical findings and therefore fills up the gap of the existent studies, by providing comprehensive explanation of the researched phenomenon.
1.2. Hypothesis
This thesis aims to focus on the investigation of a theoretical expectation that is considered the most relevant in estimating changes and trends of the agricultural production within the EU-15 member states after 2005, as this is the year when the decoupling measure was fully implemented across all countries.
The central theoretical concept regards an expectation that decoupling has a positive effect on
the agricultural production within the EU-15 member states. Therefore it is expected that
after the decoupling measure has been implemented in the CAP, the agricultural production within the EU-15 member states has increased. This assumption was based on the preliminary research findings that revealed contradictory perspectives on the influence of decoupling.
According to policy-makers, effects of full decoupling on agricultural markets are rather small, indirect and do not influence production of specific commodities. This is assumed as decoupling does not have an effect on net returns of production alternatives (Westcott & Young 2005). Therefore, this measure does not create likelihood that current production decisions would influence an allocation of financial support in the future and consequently influence an overall production output. This assumption means that a level of subsidies would be irrelevant. However, such situation is rather theoretical. In reality, the policy’s programmes do not cover all the features of these theoretical, fully decoupled payments in their objectives. Even though the main goals of this measure were to reduce trade distortions and help farmers to respond to ongoing agricultural market demands, given the policy’s complex history it is expected that there are other effects of decoupling that contradict its set objectives.
Therefore, the central hypothesis is that since the implementation of decoupling of payments
10
schemes, on the EU agricultural production has a positive effect on the production output of selected agricultural goods.
This hypothesis is built on the theoretical expectation that even though financial transfers to agricultural producers are independent from their actual production or prices of agricultural commodities, they still lead to an increase of producers’ income and their overall wealth. Despite the fact that in the policy terms this increase is presumed not to have particularly large effects on the global food production or trade, in the national scale it can have positive effects on the production. Some of these effects include a production increase of particular agricultural commodities, a development of labour, or a regional production growth in disadvantaged rural areas (Brady et al., 2011). Furthermore, as subsidies are not allocated based on the output price, producers might think that they are not facing any potential agricultural losses or risks, and thus are likely to increase their production. In general, these factors are in one way or the other affecting the agricultural output in the EU-15, and therefore confirm an assumption that decoupling measure is not production neutral but that in fact has a positive effect.
Furthermore, as decoupling is one of the CAP’s most compelling measures, it is assumed that its main objectives are focused not only on maintaining of stability but also on a continuous growth. Throughout the formation of the CAP, its main objectives substantially evolved from mainly securing sufficient food provision to European citizens to an establishment of equal and environmentally friendly agricultural procedures, primarily supporting farmers. As most of the EU budget and European policy-making have evolved around the CAP in the past forty years, it is presumed that its lengthy development and regular modifications have led to an establishment of successful, efficient and impactful policy measures, specifically the decoupling measure. Therefore, an expectation is that since 2005, when decoupling came into the effect, the measure has stimulated national economies of the EU-15 members and their overall agricultural production has increased.
The following sections will examine the previously stated hypothesis and review the existent theoretical framework in an attempt to explain the researched phenomenon. However, before the theoretical review will be established, an explicit outline of the CAP foundations will be drawn. An understanding of the policy background will help to realize the underlying factors that prompted the policy changes leading to an implementation of the decoupling measure.
11
2. Policy foundations
In order to understand what main circumstances had led to an introduction of the decoupling measure, it is necessary to examine the policy background and realize the underlying factors that prompted the major CAP changes. The assumption is that considering the policy’s prominent position on the EU agenda over the past decades, regular amendments and modernizations have led to an establishment of efficient and successful policy measures, specifically of decoupling, which stimulate the national European economies and their overall agricultural production.
The CAP was established specifically as a form of an economic regulation essentially for achieving specific policy goals in a particular economic sector. The following sections will outline periodic policy reforms, focusing on the period 2000-2016 as this duration serves as a main basis of the research. Throughout this period the most relevant legislative procedures and measures were introduced for the policy program development, including the decoupling measure.
2.1. Policy evolution
Most of the CAP's fundamental objectives have endured in its programme since the 1960s, when the Treaty of Rome established that the common European market will include agricultural goods and proposed an installation of a common agricultural policy. Central objectives of this common policy were to enhance agricultural productivity, secure fair living standards for agricultural workers, stabilize markets and guarantee products availability at reasonable prices (Snyder, 2012, p.484). In 1962 the CAP was established. Essentially, it comprised of a price and market policy as well as a structural policy on the renovation of farms.
In 1970s a focus of the CAP objectives shifted to the efficiency and an improvement of agricultural structures (Snyder, 2012, p.485). Regulatory schemes were introduced in 1987, encouraging transition from surplus to non-surplus products and reducing production output without increasing other costs. Despite the fact that the CAP attend to the entire European Community, it still had to accommodate to various national policies.
12 In the mid-1980s, main factors leading to new reforms were the policy’s increased costs, the enlargement of the Community and with it connected globalization. Direct costs of the CAP were represented in higher policy expenditure, while indirect costs resulted in higher food prices, welfare costs, health risks and environmental pollution (Snyder, 2012, p.486). The enlargement of the EU created multiple challenges as its newly joined members were substantially different from the already participating ones. Differences were visible mainly in regards to farm structures, employment, productivity, environmental and health standards. This resulted into highly difficult negotiations on the amounts of allocated subsidies, specifically of direct income payments and production limits.
In 1992, the MacSharry reforms fundamentally changed the CAP, by shifting the price support from market to producers (EC, 2016a). Support prices for the main agricultural product categories were gradually reduced, helping to increase the competitiveness of European farmers on the world markets. To avoid a decrease of producers’ incomes direct payments were established, based on the historical records of a production (either by area or livestock amounts) (EC, 2016a).
Following policy reform of 2003, so called ,Fischler reform’, originated though international pressures on the Europe to reduce domestic agricultural support as part of the 2001 Doha Development Agenda. These negotiations compelled the European Council (EC) to decrease price support, extend direct payments and transfer some costs to national governments, due to disruptive impact of the CAP on the production and global trade (Nedergaard, 2006).
The Fischler reform was the most significant in the history of the CAP, introducing a radical change of supporting farmers through the decoupling of payments from production via a Single Payment Scheme (SPS) across all EU member states. SPS was set to be paid per hectare of agricultural land and per head of livestock, but separated from the producer’s output (Brady, 2011). This scheme cut the link between income support and production. Subsidies were therefore paid aside from the fact if a farmer was generating or not generating output, provided that the land is kept in Good Agricultural and Environmental Condition (GAEC) and complied with cross-compliance rules. These were rules on public, animal and plant health, as well as environmental protection. The overall idea behind the decoupling was to move the agricultural sector towards the free market and to give farmers greater freedom to produce according to market demands. At the same time, an emphasis was shifted to stricter
13 environmental and animal welfare rules (EC, 2016a). Furthermore, the EC also permitted partial decoupling, which provided exceptions to particular member states to keep limited coupled payments on certain goods.
During the 2008 CAP Health Check further modification of the policy’ principles took place. New common directives for the direct support schemes were established in 2009 (Snyder, 2012, p.489). They expanded decoupling of direct support, increased flexibility of standards, and provided further for environmental protection.
The most recent reform for the period 2014-2020 introduced a new structure of direct payments aimed at improved targeting, equality and increased environmental protection (EC, 2016a). The CAP 2003 reform has thus been replaced by a system that links each support area to specific policy objective. The main component of this current system of direct payments, the ‘Basic payment’, remained in the form of an income subsidy (EC, 2016a). This Basic payment scheme is complemented by other support schemes, targeting other objectives or types of producers. These schemes are:
Payments to young farmers Small farmers scheme
Greening direct payment, applied for agricultural practices that are benefiting climate and environment
Redistributive payments, providing improved support to small and middle-size farms Payments for areas with natural constraints, applied for farming in difficult conditions Support coupled to production, helping sectors undergoing difficulties (EC, 2016a).
The EU member states have freedom to combine these payment schemes in order to secure effective financial aid to their agricultural producers of different categories and agricultural focuses. From the abovementioned direct payment scheme some are compulsory (including Basic payment, Greening payment, Young farmers scheme) and others are voluntary (Coupled support, Support in natural constraint areas, Redistributive payment). This means that all farmers who are qualified for support will automatically receive basic and greening payments, while others could also qualify for other optional schemes depending on an acceptance of their implementation by responding member state (EC, 2016a).
14 Looking at the evolution of the CAP policy it can be concluded that there were various internal and external factors that led to an establishment of the decoupling measure. The CAP’s objectives have shifted from the basic need for the food security to an importance of an equality and environmental protection within the EU member states. Nevertheless, since its introduction, the CAP has remained on a prominent position of the European agenda. It is therefore assumed that regular reforms have led to an establishment of successful policy measures, especially the decoupling one, which stimulate the overall agricultural production of the EU.
The following sections will examine the hypothesis stated on the beginning of the research in regards to existent academic articles and studies, in order to provide theoretical support in explaining the researched phenomenon.
3. Theoretical framework
In order to further develop and analyze the central expectation of this thesis, the existing academic studies and relevant theory in this area will be examined. The influence of the decoupling measure has been investigated by various economic researchers from both Europe and the United States (Hennessy, 1998; Wescott & Young, 2005; Becvarova, 2007; Thomson, 2009; Katranidis & Kotakou, 2010). Their studies mainly discussed plausible theoretical ways in which decoupling effects the agricultural production focusing on different areas of its impact, such as its relation to set-aside areas, productivity of farmers, their wealth and investment, sector consolidation or general production decisions and costs. These research studies used various methods of research and proved that indeed there is an influence of decoupling on the production, some indicating positive effects others indicating negative correlations. The most relevant findings will be outlined in the following paragraphs.
3.1. Impact of decoupling
Based on the existent framework, decoupling of payments can influence the agricultural production in many diverse ways and cause multiple, even contradictory, impacts. The following section will provide an overview of the most relevant studies and the results of their findings.
15
Based on a body of economic literature, which outlines various theoretical ways in which decoupling of payments has an influence on the production (Hennessy, 1998; Wescott & Young, 2005; Becvarova, 2007; Thomson, 2009; Katranidis & Kotakou, 2010), the main effects of decoupling are divided into four impact areas. These are:
1. Wealth effect 2. Sector consolidation 3. Payment basis
4. Producer risk and expectations
Theoretical findings of the studies that investigated the impact of decoupling on the production in relation to each of above-stated areas will be accordingly outlined below.
Wealth effect
In an ideal world there would be no uncertainty and risk influencing farmers’ production decisions; however in reality these factors are very real and must be accounted for. One of the first academics scrutinizing the decoupling effects under uncertainty and risk was Hennessy (Hennessy, 1998). He determined that the wealth effect of decoupling affects the production outputs under an assumption that agricultural producers are risk averse. The basic principle is that if wealth grows, farmers become less risk averse and therefore increase their production. If producers are risk averse, their actions to reduce risks or increase their earnings affect their overall production (Lopez, 2001). Farmers face uncertainty regarding the output price, so for them, differences in financial gains depend on the price variance. Decoupling policies that support producers’ incomes directly affect their decisions. Therefore, any decisions that farmers make affect both variability of produced commodities and the final production output. Consequently, based on the Hennessy’s findings, decoupling of payments influence the degree of risk aversion among producers and hence lead to an increase of their expected profits. Thus in practise, this theory indicates that decoupling has positive effects on the overall production.
16 Another research that reveals positive effects of decoupling is introduced in academic paper by Wescott and Young. Their study investigates whether decoupling leads to economic incentives that indirectly affect farmers’ production decisions, and further elaborates on the effect of decoupling in regards to producers’ wealth (Westcott & Young, 2005). As decoupled payments are not output specific their effects are mainly visible at the aggregate level, e.g. total use of land or a productivity increase. In general, aggregate decoupled payments increase the overall aggregate production. Subsequent lower prices that occur from this increase then further shape the production effects. Wescott and Young established two main effects of decoupling on the production decision-making, based on the wealth of the support beneficiaries. These are:
Direct wealth effect;
Increased investment effect (Westcott & Young, 2005).
Direct wealth effects represent an overall agricultural production growth, induced by decoupled payments that directly increase producers’ wealth. This increase indicates growth of the farm’s capital from the financing of expected future benefits for the higher land value (Westcott & Young, 2005). Production impact on the value of the land is based on the perception of the expected future support in either long term or as a single payment. Further production decisions are then established based on the land value and support it receives, which then influence the output in general. If direct payments increase producers’ wealth and thereby decrease their risk aversion level, they can lead to production decisions that raise the overall production output or change the production mix. This is usually done by switching to riskier crops that are in higher demand and have higher expected returns.
Increased investment effect caused by decoupling of payments likewise leads to an increased wealth of producers. This is occurring through the capitalization of future benefits to the land value (Westcott & Young, 2005). Growth of such investments leads to an increase of agricultural production, especially if producers would alternatively deal with debt or restricted liquidity. Producers are more likely to receive loans for the greater development of their agricultural production and operations, if they have higher guaranteed incomes and lower risk of failure. Some of these impacts are immediate and directly influence the production; others influence the output indirectly and are visible in a long term, for example by reducing production costs per unit. In general, the investment effect is positive and leads to an increased overall agricultural production.
17 Positive effects of decoupling on production were also found in a study by Serra et. al. This research examines the impact of decoupling by considering the effect that inputs have on the output variability, under the assumption that both output and output prices are uncertain (Serra, 2006). Their findings prove that price increase leads to an increase in an input use and hence raises the production output mean. Producers’ motivation to increase the production is lower if they are risk averse and if inputs are increasing possible risks. Moreover, results of Serra et. al. show that the production flexibility in regards to direct decoupled payments is positive. However, if producers are risk averse and the inputs increase risks, this positive effect disappears. The main conclusion of this study is that decoupling leads to a decrease in the variance of production output by cutting down the use of inputs, which increases possible risks (Serra, 2006).
Within the context of this thesis, it is also interesting to mention a study by Makki et.al., which analyzed the decoupling effects on farm income variance, crop choice and land allocation (Makki, 2005). Their hypothesis is built under the same consideration as Serra’s, on uncertainty about output and prices, and results prove positive effects of decoupling. The authors’ investigation indicates that decoupled payments provide only a small enrichment for producers and therefore do not significantly increase their overall welfare. Yet, this influence depends on the changing levels of market prices and on the land basis (Makki, 2005). However, their findings also prove that decoupling of payments help to protect and maintain producers’ incomes, especially during unfavourable economic conditions and in the period of low market prices. They secure that during these times not only an income will be protected but also a production output will be stimulated. If decoupled payments would not be installed, this would not be the case.
Finally, a study by Sckokai and Moro, which examines simulating of decoupling effects on cultivated land and crops, also argues that decoupling of payments is not production-neutral. This is because the positive wealth and insurance effects reimburse the negative price effect (Sckokai & Moro, 2006). Their estimated coefficients of risk aversion indicate that the larger and wealthier farms the less risk averse they are. This risk aversion represents a positive effect of decoupling, which consequently results in an increased production.
18
Sector consolidation
Sector consolidation reflects both raised productivity transfers to the non-farm economy, as well as different sizes and production structures of agricultural businesses within the EU member states. Effects of the decoupling on sector consolidation can apply on its deceleration or acceleration (Becvarova, 2007). The first effect occurs when direct payments keep inefficient and usually small businesses in function longer than they would be otherwise. Keeping these producers in business decreases the aggregate production, as the land would be used more efficiently by larger and more effective producers with higher profits. A deceleration of sector consolidation is thus caused by negative effects of decoupling.
The second effect occurs when larger businesses use the direct payments to buy out smaller ones or rent out more land (Westcott & Young, 2005). This leads to an increased aggregate production as larger businesses are better equipped and managed than smaller ones. Decoupling of payments which accelerate sector consolidation can also lead to an increased production incentives, because of the cutbacks in unit production costs. Therefore, an acceleration of sector consolidation is caused by positive effects of decoupling.
Furthermore, increased sector consolidation is considered to develop due to the larger impacts of decoupling. This estimate is based on the fact that decoupled payments are allocated unequally, by providing more subsidies to larger businesses in comparison with the overall number of producers (Westcott & Young, 2005). The main reason for this disproportionate allocation is that it is based on the size of the land and on an amount of the production output. This often leaves out smaller producers that are hence indirectly forced to exit the market.
Payment basis
According to Wescott and Young, the main basis for the allocation of decoupled support can influence expectations of producers on how the payments will be distributed in the future. Benefits linked to the past production estimations can mislead producers in believing that future payments will be linked to their historical records of the production (Wescott & Young 2005). These expectations can negatively affect current production decisions as present conditions, which apply to land or to cross-compliance guidelines, might affect the choice of
19 produced goods as well as the choice to exit the sector. In practice this means that farmers might decide to change their production preferences in cultivation of crops to the production of those commodities, for which the amount of subsidies will be higher. The whole concept of decoupling which focuses on an improved market orientation for producers would therefore deal with a situation, where producers would not be responding to market signals but to misrepresented signals influenced by the expected future profits and possible further change of conditions.
Producer risk
Increased market competitiveness and an amount of granted incentives result in a higher efficiency and an application of only relevant and effective inputs. This leads to a decrease in unit costs as ineffective inputs are consequently eliminated. In order to successfully compete in agricultural markets, farmers have to lower their production costs and supply only products that demanded by specific markets. A study by Zhu et. al indicates that income from direct decoupled payments changes producers’ time dedicated to farming, and therefore indirectly affects their production decisions and final outputs (Zhu et al., 2012). Correct production decisions are crucial as they ensure a stability of supply and directly influence both long and short term decisions on what to produce, when and how. The decoupling of financial support influences both investment and production decisions, and by that also the overall agricultural production in the long term. Another study by Findeis supports this view and confirms that decoupled payments that are supporting income do in fact lower total working time, as a result of increased possibility of the time spent at home (Findeis, 2002).
Furthermore, the most positive and direct growth effects stimulating production occurs when farmers regard their decoupled support payments as coupled, and produce on identical production levels as prior to decoupling. This is analysed in a study by Renwick and Revoredo, which points out that in the UK, in the first year after decoupling took place, output levels were essentially equal to the production levels before decoupling (Renwick & Revoredo, 2008). These production levels turned out to be clearly unprofitable and inefficient in the then current economic situation, yet producers’ decisions were still implemented. These results prove the influence of decoupled subsidies on the production output. Importance is
20 especially put on the role that producers have, particularly their perceptions and decisions, in the overall production process.
In contrast with abovementioned studies which indicate a positive decoupling effect, a theoretical study of Balkhausen and Banse suggests negative effects. It argues that decoupling of payments leads to considerable production cuts or to an increased number of voluntary set-aside areas. A Set-set-aside area is a plot of land which is taken out of production to regulate possible surpluses for which are farmers reimbursed through decoupled payments (Balkhausen & Banse, 2005). Balkhausen and Banse looked at the effects of decoupling in the EU-15, by comparing nine models which analysed the impact of decoupling in various agricultural areas and activities. Their findings showed that cereal production would significantly decrease under decoupled payments. Contradictory findings were revealed for the oilseed production estimating a growth by some models and a production decrease by others. All nine models indicated that beef and sheep meat production are projected to decline, but the models vary in what is significance of the decline level (Balkhausen et al, 2008). Furthermore, the current decoupled direct aid framework allows producers to obtain financial support even without occurring production costs. This results in farmers discontinuing their production, for example by setting aside their land or completely exiting the market.
Consequently, after analyzing some of the most significant theoretical framework indicating that the possible effects of decoupling on the production are mainly positive, it seems reasonable to assume that positive effects of decoupling overweight the negative ones. This is established from the review of the abovementioned theoretical frameworks that provide multiple ways in which decoupling increases production output in one way or another. This supports the central hypothesis that the decoupling measure has positive effects that stimulate the overall agricultural production.
Even though the impact of subsidies on the agricultural production after the implementation of decoupling, had been documented in the academic literature from the theoretical point of view, comprehensive empirical research has not been undertaken after the most recent CAP reform 2013, thus leaving space for the hypothesis testing. The following section will outline what research design will be employed in the analysis, and introduce what variables and data will be used for the empirical examination of the hypothesis.
21
4. Research design
For the purpose of answering the central research question, the research strategy employed for this thesis will be panel data regression analysis. This method allows for more valid and reliable responses compared to a case study or an experiment. The main findings from theoretical framework indicated a relevance of a relation between decoupling of agricultural subsidies and the agricultural production. These theoretical results indicated that impacts of decoupling are both production inducing as well as deteriorating. The analysis will test these theoretical propositions as well as the established central hypothesis, and either confirm or reject them.
Therefore, quantitative research method will be implemented for the data analysis, by employing an econometric technique using regression with panel data as a main measure. Panel data allows for observations of the same entities to be examined at multiple time periods. It helps to control for omitted variable bias or unobserved, immeasurable sources of various individual heterogeneity that does not vary over time (Hsiao, 2014). Observations that cover both time and individual units cross-sectionally allow gaining more information and more efficient estimates. Regression will be run over both dimensions in order to determine the strength of a relation between dependent variable and a series of other independent variables that are changing.
The data set that will be used for the observations covers financial period of 2000–2016, as decoupling measure was fully implemented from 2005, and will include all member states of the EU-15. Therefore, the impact of the decoupling will be analysed by comparing its influence on agricultural production levels within the EU-15 countries before benefiting from this measure and after its implementation. The assumption is that, after the decoupling measure has been implemented in 2005, the influence of the CAP subsidies was stronger and more effective hence increasing the overall agricultural production levels.
In regards to the feasibility of this research it is out of the scope to take into the account all external and internal factors that are directly influencing agricultural production in every member state of the EU-15. Therefore, the research included only the main external factors that further considered as control variables. Nevertheless, the analysis of the selected data within the limitations provide valid and credible insights into the production output across the
22 EU-15 and serve as sufficient basis for realizing the decoupling impacts. For the reference, the analysis will use data from the first pillar, namely the Decoupled direct aid (Article 05 03 01) covered in the Basic payment scheme of the CAP’s direct payments.
4.1. Operationalization
In order to transform the main research question into researchable themes that will be suitable for the empirical research, abstract concepts from the theoretical framework need to be operationalized.
The main research question is: ,,What is the influence of decoupling of direct payments on the
agricultural production within the EU-15 member states with regards to the output of their agricultural commodities?’’
Main variables will be conceptualized to measure the effect of subsidies on the agricultural production after the decoupling was fully implemented in 2005. The analysis looks at the effect on both the production in total as well as on the production of specific agricultural commodities, either plant or animal based. To operationalize key regression variables, both dependent and independent variables are established. For dependent variable, the total production is measured in millions of EUR (its values), for the production of specific agricultural commodities in volumes (1000 t). The independent variable is represented by the decoupled direct aid in million EUR.
Dependent variable: Production Independent variable: Decoupling
Control variables: GDP, inflation, population Indicators:
Difference between production trends before the policy change (year 2005) and after the change
Level of the impact of subsidies on the total production output after an implementation of decoupling
Level of the impact of subsidies on plants and livestock production output respectively after an implementation of decoupling
23 Regression coefficients indicate a change of the dependent variable from an impact of the independent variable. The regression equation, which is used to draw conclusions on whether the impact of subsidies on the production was stronger after the policy change in 2005 or before it, is thus:
prod= α + β1 Expt + β2 Dt + β3 Dt Expt + control variables + εt
prod - Dependent (outcome) variable Expt - Independent (treatment) variable
Dt - Dummy variable
Dt Expt - Dummy for the period of the policy change
D=0 for the period before the change: 2000-2005 D=1 for the period after the change: 2005-2016
Parameters:
α - Constant
β - Treatment effect εt - Error term
In the regression equation, the variable β1Expt represents a general effect of subsidies on the
agricultural production in the period 2000-2016. Negative coefficient indicates that higher expenditure causes on average a lower agricultural production. As the main question examines if the influence of subsidies before decoupling (year 2005) is different than after decoupling, the most relevant variable that represents this relation is β3DtExpt. This is
illustrates the difference between production trends before and after the change. If the effect is positive, then the impact of subsidies after the change is greater than it was before the change. Thus:
if β3 > 0, then result is consistent with the hypothesis,
if β3 < 0, then result is inconsistent with the hypothesis.
The dummy variable helps to establish if an observation has or has not have an effect that changes the outcome. In this case, it is defined as 0 before the introduction of direct aid (before 2005) and 1 after 2005. This means that before the introduction of the decoupling of
24 subsidies there was no effect of decoupling on the agricultural production, and after 2005 there was an effect, as measured by the dummy variable.
To obtain an unbiased estimate of the tested effect in the regression, it has to be established that the coefficient on the independent variable does not suffer from the omitted variable bias. That is why the control variables will be added to the regression as their purpose is to eliminate this omitted variable bias. The first control variable that will be used in the regression is inflation. This variable will control for an increase in the direct aid support due to the rise in consumer prices, as well as to control for the effects of price changes on the agriculture production. The second control variable in the regression will be GDP, to control for potentially confounding factors correlated with the agricultural production. It is expected than if the GDP is rising than it is likely that there is a higher demand for agricultural products, leading to an increased overall production. Third control variable that will be used in the regression is population, as there are different population sizes among the EU countries that together with the population growth influence the demand, resulting into the higher agricultural production.
4.2. Data and limitations
To carry out the empirical data testing of the hypothesis, certain limitations have been imposed to make the data gathering feasible within the given time-frame. As the theoretical expectations hypothesize on different effects of the decoupling for variation of the EU countries, for the purpose of this thesis the original EU-15 structure has been selected. The EU members vary in many ways, with some countries being part of the EU since the very beginning and others joining as late as 2013. This has influenced the national implementation of the CAP in the agricultural structures of member states, and presumably their subsequent overall production outputs. For the EU-15 countries the data on amounts of subsidies and production trends were sufficiently accessible, and as these countries were affected by the CAP from its inception, they were chosen for a basis of this analysis. In the analysis a selection of 15 countries will represent 240 observations.
25
Production data
To establish the effect of decoupling on the production, data on the production outputs and volumes of particular agricultural commodities, covering categories of both plant and animal based goods, will be accessed from the database of Eurostat (Eurostat, 2016). The analysis will examine the impact on the production for specific selected agricultural commodities to observe the possible effect of decoupling in greater details. Selected commodities will be chosen based on their relevance on the market and an availability of the data, to allow for a feasible research. The data availability varies significantly from country to country and from product to product. For instance, there is no data available for certain countries on the production of particular goods up until a certain years, such as Portugal and Germany in regards to vegetables. However, since this database provides comprehensive data available for majority of years, products and countries, it is the most relevant source for the overall analysis.
Decoupled direct aid data
The amounts of the decoupled subsidies allocated to each member state will be accessed from the EC’s financial reports, which describe the CAP funding for period 2005-2016 (EC, 2017a). Decoupled payments are not output specific, and thus influences are more visible at the aggregate level. Therefore, the detailed outline of information on the budgets allocated to each specific agricultural commodity for the individual states of the EU-15 is not available. Consequently the analysis will use only aggregated data.
Data on control variables
The data on both inflation and population for all of the EU-15 countries will be likewise obtained from the Eurostat database. In order to measure the inflation the harmonised index of consumer prices (HICP) will be used, which is the consumer price index as calculated in the EU, based on the harmonised approach and a set of guidelines across all member states. Furthermore, GDP data for all EU countries will be gathered from the UN database, and collected in euro value with 2005 as index year.
26 The main challenges with the data analysis is its availability, as despite the fact that the used database provides a majority of needed information there are still some missing data on the production outputs for particular agricultural commodities or years. In regard to the feasibility of this research it is out of the scope to take into the account all external and internal factors that directly and indirectly influencing the agricultural production in every state of the EU-15. Such examination would require longer time-frame of the research and detailed in-depth analysis of vast amount of information that is rather difficult to isolate, such as variety of direct payment combinations between mandatory and voluntary schemes, individual national adjustments of the policy for every EU-15 state, market conditions, etc.
Therefore, an absolute and accurate assessment of the impact of decoupling across the EU is rather difficult, because of the staggered nature of its implementation across different countries and the complexity of capturing impacts of other key drivers. These factors make the quantification of the actual impact challenging, as they complicate the identification and reliable estimation of both direct and indirect effects. For more accurate and in-depth estimates of the effect of the decoupling on the national agriculture of EU members, production levels per capita or production as a share in national income should be considered, however these were out of the scope of this research and thus will not examined.
4.3. Production trends
To understand the influence and changes in production which the established regression equation will reveal in the analysis, it is important to understand what was the observed situation and trends in the agricultural production over the investigated period 2000-2016. Eurostat database provided detailed information on the agricultural data for each of particular commodities relevant for the research. In order to implement selected research method that employs regression equation, it is necessary to clearly define what the agricultural commodities are, either crop or animal based, which are investigated throughout the research. These will be defined in following paragraphs.
The term ‘crop’ comprises a wide spectrum of cultivated plants that are being grown in agriculture. The variety of crops across the EU reflects the topographic and climatic conditions of their environment, as well as pests and diseases. In the context of this paper, production of crops that are harvested for the consumption purposes for humans or livestock
27 will be scrutinized (Eurostat, 2017). Industrial non-food crops will not be considered in the analysis. The analysis of results will be divided according to the main agricultural categories which were cereals, fruits, vegetables and animal based goods.
Category of cereals include wheat, rye, maslin, barley, oats, mixed grain, grain maize, sorghum, triticale, buckwheat, millet, canary seed and rice (Eurostat, 2013). Aside from this category, which covers multiple different crops, the analysis will also examine the impact of decoupling on the production of several individual crops. These will be wheat, rye, barley, oats, rice, potatoes, oilseeds, rape and tobacco. Wheat, rye, barley and oats have been selected to be analysed separately, because they accounted for the highest share (86%) of the cereals produced in the EU-28 in 2015 (Eurostat, 2017). For a category of wheat, the regression equation established in this paper will analyze data which will comprise of data on common wheat, spelt, einkorn wheat and durum wheat. For a category of rye, data on rye, mixtures of rye and other cereal mixtures sown before or during the winter (maslin) will be included. Category of barley contains data on winter barley sown before and during winter, as well as spring barley that is sown in the spring. Category on oats comprise of oats and other cereals sown in the spring, grown as mixtures and harvested as dry grain, including seed.
Below is a demonstration of production trends of cereals over the period 2000-2016 (Graph 1). This representation reveals that the top three producers of cereals in the EU-15 are France, Germany, and United Kingdom. France and Germany are without a doubt the largest cereal producers, France with an average production share of 65 274.87 million tonnes and Germany with an average harvested production of 46 191.59 million tonnes in the period 2000-2016 (Table 1). As the decoupling took place in 2005, it is interesting to compare production trends before 2005 and after, when decoupling was implemented. For the four biggest cereals producers indicated in the graph 1, harvested aggregated cereals production has on an average increased after a year 2005. Two exceptions are Italy and United Kingdom. For Italy aggregated production turned out to be on average 1448.49 million tonnes lower after 2005 than it was before. For United Kingdom this difference was smaller, indicating on average 38.58 million tonnes decrease in production after 2005 than before it (Table 1).
Furthermore, graph 2 illustrates production trends of wheat and spelt (Graph 2). It can be seen that the top three producers are likewise France, Germany and United Kingdom. This confirms the statements above that these three countries indeed dominate cereals market in the
28 EU-15. Comparing production trends before and after 2005, for Denmark, France, Spain and Germany aggregated cereals production has on an average increased after 2005. Production for Italy and United Kingdom remained higher before 2005 (Table 1).
0.00 10,000.00 20,000.00 30,000.00 40,000.00 50,000.00 60,000.00 70,000.00 80,000.00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Graph 1. Cereals
Denmark France Spain Germany United Kingdom Italy
0.00 5,000.00 10,000.00 15,000.00 20,000.00 25,000.00 30,000.00 35,000.00 40,000.00 45,000.00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Harve ste d p rod u ction (t ) Year Graph 2. Wheat and spelt
29 Table 1.
Mean for cereals production Mean for wheat production
Countries 2000-2016 2000-2005 2006-2016 2000-2016 2000-2005 2006-2016 Denmark 9,173.49 9,156.12 9,182.97 4,763.88 4,626.77 4,838.66 Germany 46,191.59 45,812.50 46,419.04 23,589.10 22,276.13 24,305.27 Spain 20,343.51 19,923.82 20,572.43 6,309.28 6,044.32 6,453.81 France 65,274.87 64,085.98 65,923.35 36,595.26 35,814.42 37,021.18 Italy 18,438.54 19,375.80 17,927.31 7,316.97 7,328.98 7,309.76 UK 21,746.85 21,771.82 21,733.24 14,755.47 14,819.67 14,720.45
Moreover, the analysis will examine decoupling effect on the production of potatoes separately, as they are one of the main root crops in the EU-28 producing 53.1 million tonnes in 2015 (Eurostat, 2017a). As this number has decreased in the past five years it was chosen as relevant to be examined in the analysis. Oilseeds were included in the analysis, as in the recent years a growing use of oilseeds for bio-energy production supposedly boosted its demand (Eurostat, 2017a). Category of rape comprises of data on rape, turnip rape, sunflower seeds and soya, harvested as dry grains. In 2015, the production of rape exceeded the average of previous years, reaching 21.7 million tonnes in the EU-28 (Eurostat, 2017).
One of the key components in the EU agriculture is the fruit sector. Data regarding fruit that will be examined in the analysis include apples, pears, stoned fruits (e.g. peaches, apricots), nuts, other top fruits (e.g. figs or kiwi), other soft fruit that grows on bushes (e.g. berries, currants) and wild fruits. Vegetables cultivated for fruit are also included in the fruit category; these are tomatoes, cucumbers, gherkins, melons, aubergine, pumpkins and red pepper). Separately from the overall fruit category, individual categories of grapes, citrus fruits, strawberries and olives will be established and the impact of decoupling on their production will be scrutinized. As the EU is the world's leading producer of wine, the wine market is highly developed and significantly influences the demand for grape production. In regards to the CAP, the 2013 reform replaced the EU planting rights by a scheme of authorisations for vine planting and enabled competitive producers to increase their production (Eurostat, 2017). That is why it is interesting to analyze the individual effects of decoupling on the production of grapes. Citrus fruits and strawberries were established to be analysed separately as they are were not included in the statistical data of the category comprising all fruits, because of their
30 distinctive characteristics and classification. Since the EU is globally the largest producer of olive oil supplying three quarters of global production, it has been also chosen to be scrutinized separately in greater details (Eurostat, 2017).
Graph 3 demonstrates production trends of fruits in the EU-15 over the period 2000-2016 in greater details. In terms of economic significance Italy, Spain and France are on average the most important fruit producers (Graph 3). Italy is on the lead, as it has generated on average 6,020.29 million tonnes in the period 2000-2016 (Table 2). Even though some data for certain years were not available in the databases it is apparent that the production for these has still followed an upward pattern, which is indicated in the high data from the latest available years. After an implementation of decoupling measure in 2005, the production of fruit on average has signified an increase across all of the main six producers.
Trends in the production of grapes in the EU-15 over the period 2000-2016 have been illustrated in the graph 4. As European wines are highly favoured on global markets, it is interesting to observe what impact, if any, did CAP reforms of 2005 and 2013 have on average for the production of grapes. Trends data indicate that after 2005 grapes production has decreased in Greece, Spain, France and Portugal. On a contrary, in Italy and Germany the aggregated harvested production has on average resulted higher after 2005 than in the period before (Table 2).
Table 2.
Mean for production of fruit Mean for production of grapes
Countries 2000-2016 2000-2005 2006-2016 2000-2016 2000-2005 2006-2016 Belgium 581.50 556.92 606.08 - - - Germany 1,137.24 1,108.32 1,154.59 1,289.83 1,281.08 1,302.95 Greece 1,449.58 1,356.17 1,500.53 1,047.12 1,156.26 987.59 Spain 3,743.70 3,971.40 3,553.96 6,124.64 6,201.54 6,089.68 France 3,224.43 3,274.29 3,197.23 6,416.11 7,049.66 6,070.55 Italy 6,020.29 5,776.08 6,313.34 8,333.52 8,274.08 8,365.94 Portugal - - - 888.83 975.05 841.80
31 0.00 1,000.00 2,000.00 3,000.00 4,000.00 5,000.00 6,000.00 7,000.00 8,000.00 9,000.00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 H a rv este d pro du ct io n (t ) Year
Graph 3. Fruits, berries and nuts
Germany France Greece Spain Italy Belgium
0.00 2,000.00 4,000.00 6,000.00 8,000.00 10,000.00 12,000.00 14,000.00 16,000.00 18,000.00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 H a rv este d pro du ct io n (t ) Year Graph 4. Grapes
32 Other essential component of the EU agriculture is the vegetable sector. In 2015, the most important vegetables producers were The Netherlands, Spain and Italy, producing over 60 % of the overall EU-28 fruit output (Eurostat, 2017). Hypothesis-testing analysis of this paper will focus on three main categories of vegetables. These include category of brassica (e.g. cabbage, cauliflower and broccoli), category of leafy or stalked vegetables (e.g. celery, lettuce, spinach and asparagus), and category of other fresh vegetables that include root and tuber vegetables (e.g. turnips, carrots, onions, garlic, beetroot and radishes) and pulses such as peas and beans (Eurostat, 2012).
Trends for production of vegetables in the EU-15 are presented in graph 5 and graph 6. Graph 5 reveals trends for the production of fresh vegetables, graph 6 trends for the production of leafy vegetables throughout the period 2000-2016. Major vegetable producer is Italy, which over the analysed period cultivated on average 13 397.73 million tonnes of fresh vegetables and 2438.48 million tonnes of leafy vegetables (Graph 5; Graph 6). Second dominant player in the market is Spain, producing on average 12 571.37 million tonnes of fresh vegetables and 1469.99 million tonnes of leafy vegetables. After the policy change in 2005 it becomes evident that on average the production of fresh vegetables has increased for Spain, Germany, The Netherlands (Table 3). This can be observed despite the fact that some data were not available for certain years, as the growth path indicates a direction of the trend. Other main producers of the EU-15 experienced on average higher levels of vegetable production before 2015 (Table 3). Regarding the production of leafy vegetables, on average increased production after 2015 was revealed only for Germany and Italy.
Table 3.
Mean for fresh vegetables Mean for leafy vegetables
Countries 2000-2016 2000-2005 2006-2016 2000-2016 2000-2005 2006-2016 Germany 3,450.36 3,012.20 3,490.19 603.93 394.50 622.97 Greece 3,538.35 4,014.32 3,278.73 - - - Spain 12,571.37 13,003.57 12,139.17 1,469.99 1,795.12 1,144.87 France 5,743.08 6,346.85 5,380.81 1,116.79 1,385.28 886.66 Italy 13,397.73 14,174.17 12,815.40 2,438.48 2,409.48 2,473.28 Netherlands 4,370.50 4,182.80 4,472.88 450.27 520.60 411.91 Belgium - - - 882.36 1,115.88 682.20
