What will be the effects on the production of milk in the
Netherlands when the milk quota will be abolished in 2015?
Author: Jeroen Sanderink
Studentnumber:10211411
Specialization: Economie en Financiering
Faculty of Economics and Business
University of Amsterdam
Name of the evaluator: Lin Zhao
31 December 2014
This research aims to sketch the future production process in the Dutch dairy industry after the milk quota abolishment. The first part gives an overview of the quota within the
Netherlands, and the differences between other milk quota systems. In the second part, two models are developed to determine the importance of the explanatory variables. The
explanatory variables and the projections of the explanatory variables are based on the literature studies and some logical reasoning. The first model tries to predict the impact of the quota on the future milk production, the second model on the price. There are six scenarios analyzed. The results indicates that the quotas’ impact on the production could result in increased production and lower milk prices. The overall effect for the Dutch dairy production, according to this thesis, will be positive. At the end of this thesis, propositions are given for further research.
2
Index
1. Introduction ... 3
2. Literature Review ... 5
2.1 Introduction of the quota ... 5
2.2 Implementation and mechanism of the milk quota ... 6
2.3 Theoretical effect of a quota ... 7
2.4 Inefficiencies of the quota system ... 8
2.5 Consequences of the current regulations ... 9
3. Method and Data ... 11
3.1 Variables and data sources ... 11
3.1.1 Demand-related variables ... 11
3.1.2 Supply-related variables ... 12
3.2 Milk output changes due to the changing production inputs ... 14
3.3 Milk price changes due to the abolishment of the quota ... 14
3.4 Possible effects of the quota’s abolishment after 2015... 15
4. Findings... 16
4.1 Results of the models ... 16
4.2 How the milk output change ... 18
4.3 How the milk price change ... 19
4.4 How would future production develop ... 21
5. Conclusive Remarks ... 22
Reference List ... 24
3
1. Introduction
During the late 1970s and early 1980s there were problems due to excess milk production in the European Union that would led to decreases in price. But at that time the European Union wielded a price support for dairy output. This meant that the Member States were obliged to give a subsidy on dairy products, when the prices declined below the agreed intervention price. It was an intervention strategy by countries of the European Union in order to assure a relative income position for farmers (Bouamra-Mechemache et al., 2008b). Because the costs of this policy rises rapidly during that time, policy makers introduced a temporary milk quota in 1984 (Parton, 1992; Boots et al., 1997). The plans were to abolish this quota within 5 years (Parton, 1992).
In the current situation the milk quota is still active. Some changes have been made over the years, with regard to the quota per member state and the manner how it is tradable
between farmers within a Member State. But the main purpose of the milk quota still provides a limit of the total milk production to keep milk prices above the free market level (Wieck et al., 2007). If producers produce more than they are entitled to, they will be punished by a levy (Oskam et al., 1992).
The plans are to abolish the quota in 2015, because there are still obstacles to quota trade and inefficiencies in the design of the quota market (Oudendag et al., 2014; Wieck et al., 2007). To create a gradual transition to a quota-free system, the European Commission applied a total quota increase of 1.5 % in 2006/2007 (Samson et al., 2012). In 2008, the total quota was increased by another 2 % and after 2009/2010 the total quota was increased by a 1 % per year until 2013 (Bouamra-Mechemache et al., 2008a; European Commission, 2009; Samson et al., 2012).
A quota isn’t binding for every Member State of the European Union. Some member states have structural lower milk production than the milk quota permits (Oudendag et al, 2014; Wieck et al., 2007). This is not the case in the Netherlands. The Dutch dairy farmers are producing at the maximum quota levels, so quotas for them are still binding (Samson et al., 2012). This led to less efficient production (Boots et al., 1997).
The Dutch farmers are expanding their means of production to benefit for the changes after 2015. They are building greater cowsheds and invest in more advanced machinery, in order to meet future developments (Samson et al., 2012). These expansion strategies should led to efficient production facilities in the future so that they can eventually produce more milk. I find this development so interesting to focus my research topic on this. It is very interesting to investigate if this is a positive trend, or that the Dutch farmers are throwing money down the drain.
Earlier studies on the abolition of the milk quota performed by Bouamra-Mechemache et al. (2008b); Chantreuil et al. (2008); Helming & van Berkum (2008); Lips et al. (2005); and Samson et al. (2012) all assumed that there is a large potential of milk production increase when milk quotas are abolished. According to these studies it is highly likely that the milk price will decline.
Another effect of the abolishing of the quota, according to Oudendag et al. (2014) and Wieck et al. (2007), is that small farmers will be crowded out of the dairy sector. Their yields are too low to proceed their production as a result of declining milk prices.
4 The research question of my thesis is: “What will be the effect on the production of milk in the Netherlands when the milk quota will be abolished in 2015?”. In order to answer my research question, I will first conduct a literature review of the quota system in the European Union and especially in the Netherlands. After that I will conduct an empirical analysis
exploring the formation of milk production and milk prices in the Netherlands.
The literature review will deal with differences between the Europe- and Dutch dairy sector. When the differences are known, it is easier to see which country will most likely expand their production, or absorb another Member States’ production after the
abolishment. The literature also helps to determine the most important variables that explain the establishment of the milk production and the milk price.
In the empirical analysis I will introduce two models. One model to explain milk
productions and the other to explain milk prices both for the Netherlands. Projections will be made for the explanatory variables to get future projections for the milk production and milk price. When the future milk productions and future milk prices are predicted by these
models, the price effect and production effect for the Netherlands will be analyzed to give a final conclusion to the effect on the production in the Netherlands when the quota will be abolished.
According to the results of the empirical analysis, milk price decreases as a result of increasing milk production. This is also predicted by the studies of Helming & van Berkum (2008); Lips et al. (2005); and Samson et al. (2012). I think that, based on the high quota price and the binding quota for the Dutch dairy sector, it is highly likely that the Netherlands will expand production more than the average production expansion in the European Union (Samson et al., 2012). Helming & van Berkum (2008) and Lips et al. (2005) also came to the conclusion that the production in the Netherlands will increase more than in most other countries. Therefore I conclude that the increase in production will highly likely have a greater impact for Dutch farmers, than the decrease in price would have.
My thesis proceeds as follows. The next part contains a brief literary composition about: the introduction of the quota; implementation and mechanism of the quota system;
inefficiencies of the quota; and recent developments in the Dutch dairy sector. In the third part of this thesis the research models are developed and the origin of the data that I used are explained. The findings that I have obtained are presented in the fourth part. And in the last part some conclusive remarks are given, as well as recommendations for further
5
2. Literature Review
This chapter starts with mentioning the period antecedently of the introduction of the quota. After that, a theoretical framework of the milk quota and its implementation mechanism is discussed. In the subsequent paragraph the theoretical effect of a quota is briefly reviewed. Thereafter, the inefficiencies of the quota system, and how it is been solved, is explained in greater detail. In the last paragraph, the significant changes of the European regulation for the Dutch dairy sector are discussed. I also state my hypothesis on these changes, and their likely outcome at the end of this paragraph.
2.1 Introduction of the quota
In the period preceding the introduction of the quota system there were multiple regulations used by the European Union in the dairy sector. In order to keep a relative income for farmers, the European Union used intervention purchases, subsidies on
consumption and exports, and tariffs on the import (Wieck et al., 2007; Parton, 1992). The costs of these regulations were one of the main problems (Parton, 1992). Member States were obliged to give a subsidy on dairy products, when the prices declined below the agreed intervention price. Therefore the costs to manage the support prices rise exponentially when the surplus of production rises. This has happened from the late 1970s into the 1980s
(Parton, 1992). In order to assure a relative price for producers and lower the regulation costs, the European Union agreed to implement a quota system on dairy production. The main reason for this quota system was to reduce production so that the producers receive higher prices for their milk and therefore a higher income, without additional costs for Member States (Oskam et al., 1992).
After the introduction of the quota system, the cost of The Common Agricultural Policy budget (CAP-budget) declined drastically. The CAP-budget was around 70 % of the total European Union expenditures in the beginning of the 1980s (European Commission, 2014). The quota system had lowered the support costs, and the Common Agricultural Policy expenditures declined to approximately 55 % by the end of the 1990s (European
Commission, 2014; Graph 1.1). Whereof 6.7 % of the CAP-budget was for the dairy sector and 40.6 % was for the arable crops (European Commission, 2000). Arable crops are used to feed cows in the dairy sector. After the introduction of the quota system the size of the quota per country has been adjusted a few times, but for the EU-15, the peak of 1983 is never reached again (DiaryCo.org.uk, 2014a).
6
Graph 1. 1 CAP expenditure: European Commission, DG Agriculture and Rural Development (Financial Report).
2.2 Implementation and mechanism of the milk quota
To allocate the initial milk quota for each Member State, a reference period was selected based on previous years of production (Wieck et al., 2007). The quota provided a limit to production, and crossing this limit would result in a penalty for producers (Colman, 2000; Oskam et al., 1992).
Because milk production is limited by the quota, milk prices would be relatively higher. This is positively reflected in the income for farmers. But farmers have to cope with costs that wasn’t previously at issue, namely quota prices (Boots et al., 1997; Colman, 2000; Oskam et al., 1992). These quota prices are suppressing their additional profits (Boots et al., 1997; Colman, 2000; Oskam et al., 1992). Paragraph 2.3 provides a theoretical approach to the cost of quotas.
However, not in every Member State farmers are facing the same quota costs. Oskam et al. (1992) state that these differences arise by the type of quota market a government uses for the structural development in the presence of a quota. There are two extremes that can be distinguished: an administrative quota and a marketable quota.
First, a government can regulate the quota by administrative regulations so that the government will decide which farmer will be appointed to get a quota. This approach pursues special national objectives linked to rural, social, or economic development. (Wieck et al., 2007; Oskam et al., 1992). For instance the government can give quotas to farmers in less-developed and mountainous regions, or it can help new entrants or smaller farmers to expand (Oskam et al., 1992).
The other regulation type of a quota system is through the so-called marketable quotas. It is allowed by the European Union to trade marketable quotas within national borders. Quotas in a quota market have, in theory, a price, which reflect the difference in milk price and the economic profit of alternative use of the production inputs (Oskam et al., 1992).
7 With a marketable quota, the efficient producers with low marginal costs are willing to pay more for a quota than lower efficient producers. Hence, the quotas will be transferred to the most efficient producers (Wieck et al., 2007; Oskam et al., 1992). And therefore is quota trade more applied by the larger farms than by the average farmers (Oskam et al., 1992). This occurs because they have better access to finance, and experience economies of scale (Oskam et al., 1992).
Some Member States used exit schemes. Through exit schemes, existing producers who had been allocated a quota could choose during the first year of the regulation to exit the market and in return get a compensation (Oskam et al., 1992; Wieck et al., 2007). These exit schemes existed to ensure further and faster structural development at the beginning of the quota (Oskam et al., 1992; Wieck et al., 2007).
The Netherlands also introduced exit schemes to accelerate the structural development in the first year of the quota system. Oskam et al. (1992) state that these exit schemes were at first generally successful, but over time the market price of a quota became higher than the compensation of the exit schemes. Therefore it was difficult to find producers willing to quit their dairy production in order to transfer the quota to more efficient producers. In contrast, the France exit schemes were very successful in reducing the number of dairy herds. The difference is due to the market type that is used. France use the administrative
allocation, and the Netherlands manage a marketable quota. The result is that a quota
market is incompatible with exit schemes (Oskam et al., 1992).
2.3 Theoretical effect of a quota
The partial equilibrium competitive model states that the equilibrium price of a good depends on the interaction of supply and demand (Snyder et al., 2012). When supply increases, ceteris paribus, the supply curve will shift to the right, this is illustrated in figure 2.1. The result is that the price of the good will decline.
A quota system distorts this interaction, and results in higher profits for producers. The quota system sets a limit on production for each Member State (Lips et al., 2005). As can be seen in figure 2.2, the price consumers want to pay is therefore higher with a quota, QQ,
then without a quota, Q, but with fewer transactions made. The extra profit a quota holder receives for the traded goods is called the quota rent, and is indicated as the blue marking in figure 2.2 (Krugman et al., 2012). This extra profit turns out lower in a marketable quota
system, because the producers have to acquire a quota first, and this quota has a price.
8 The dead weight loss is the red marking illustrated in figure 2.2, consisting of mutually
beneficial transactions that are discouraged by the quota (Snyder et al., 2012). The costs of a quota can be faced as a loss in consumer and producer welfare. The benefit of a quota in contrast to a subsidy is that the costs of a quota are only faced by the consumer and producer and not the government, in this thesis the Member States.
Farmers are making positive profits due to the quota rents. Each farmer wants to produce against the lowest possible marginal costs, in order to obtain the highest possible quota rent (Samson et al., 2012, p. 2). When production increases, the quota rents will decrease due to higher marginal costs for farmers (Samson et al., 2012). The potential increase in milk production when milk quotas are abolished is large according to Samson et al. (2012).
2.4 Inefficiencies of the quota system
Marketable quotas ensures that farmers produce more efficiently. This is because most
efficient farmers are willing to pay the most for a quota (Oskam et al., 1992; Wieck et al., 2007). However, due to quota trade restrictions, the quota trade market was distorted and has not always be efficient. For example, some farmers wanted to produce more than the quota permits in a certain year, caused for instance by favorable weather conditions in their region. They wanted to avoid a super levy, and therefore wouldn’t produce as much as their inputs can manage. These inefficiencies manifest as costs for the producer and slow down structural development (Boots et al., 1997; Oskam et al., 1992).
To ease the rigidities of the quota system, the quota system was subject to change. At first, the quotas were only transferable through the sale of land (Boots et al., 1997; Oskam et al., 1992). In 1989, the Netherlands adopted leasing of quotas to achieve more economic efficiency (Boots et al., 1997). Because milk quotas were tied to land, farmers whom used the quota rights were required to buy or rent the land for at least a year (Boots et al., 1997). Boots et al. (1997) states that the leasing of quotas was restricted to a maximum of 20 thousand kilograms of milk per hectare. According to the authors, when suppliers wanted to lease out milk, there had to be a minimum transfer of 10 thousand kilograms of milk.
Moreover demanders might not lease more than 75 thousand kilograms of milk.
The leasing instrument gives producers more flexibility in planning one year’s production, and thereby not risking a super levy (Oskam et al., 1992). The market efficiency increased, because the financial barriers for farmers became lower, and the mobility of quota transfers increased (Oskam et al., 1992). A drawback of the quota leasing is that the exchange of both ownership and user rights incurs additional transaction costs, like notary and estate agent costs (Boots et al., 1997). The notary and estate agent costs are estimated by Boots et al. in 1997 at 2% of the milk quota price.
Since 1990, the inefficiencies of the marketable quota are almost nothing, because producers can split their land and quota into separate tradable goods and thereby is a quota freely tradable in the Netherlands (Oskam et al., 1992). The United Kingdom also manage a almost free milk quota market (Wieck et al., 2007). The other Member States, like Germany and Denmark, can be ranked between the free marketable quota approach and the
9 2.5 Consequences of the current regulations
The dairy sector in the European Union is facing significant changes through regulation. It is important to analyze these changes, because the regulations have a causal impact on the production and the prices (Oudendag et al., 2014; Parton, 1992). In this paragraph I give assumptions regarding further developments in the Dutch dairy sector.
In 1999 the EU 15 countries agreed on the Agenda 2000 Reform that gives several changes in dairy production. The countries decided to extend the milk quota system until 2008 (Lips et al., 2005). In order to cater to the problems with the artificial quota shortage and related high costs faced by farms that wanted to expand, led to the decision to increase the total quota volume instead of further decreasing it (Wieck et al., 2007). The Agenda 2000 have also brought considerable declines in the market price support for the EU dairy sector (Chantreuil et al., 2008). The effect of the decreasing support prices is that the milk prices became more volatile, since the bottom prices were lowered (Samson et al., 2012).
Second there was the Luxembourg reform that became active in 2004
(Bouamra-Mechemache et al., 2008b). This resulted in lower support prices for several end products of milk. They further decided to extend the milk quota system until 2015, with a 0.5 % increase in 2006 and 2007 (Bouamra-Mechemache et al., 2008b; Samson et al., 2012).
At last, the European Union endorsed a proposal of a 2 % increase of the milk quota in 2008 and from 2009 until 2013 a gradual increase of 1 % (European Commission, 2009; Samson et al, 2012). The data shows that all the EU 15 countries increased their milk
production from 2008 (DairyCo.org.uk, 2014a). Finally, there will be a complete abolishment of the quota in the European Union in 2015 (Samson et al., 2012). The projections that will be made for the explanatory variables are based on the years after 2008, so that the projections will partly anticipate the quota changes. The projections of the explanatory variables are explained in greater detail in the next chapter.
Because the dairy markets are characterized by inelastic demand and supply, a relatively small change in the demand or supply of milk can result in a large impact on the milk prices (Samson et al., 2012). The effect of these quota expansion policies are therefore likely to result in lower milk prices.
Earlier studies on the abolition of the milk quota performed by Bouamra-Mechemache et al. (2008b); Chantreuil et al. (2008); Helming & van Berkum (2008); Lips et al. (2005); and Samson et al. (2012) all assumed that there is a large potential increase of milk production when milk quotas are abolished. Likely resulting in lower milk prices according to these authors.
According to Samson et al. (2012) and Helming & van Berkum (2008), removal of the quota system in the European Union will respectively result in an overall 11 % to 21 % more milk production in the Netherlands. Chantreuil et al. (2008) and Lips et al. (2005) argued that the EU-wide milk production increase is respectively around the 3 % and 4 %. This means that the Dutch dairy sector has a comparative advantage in comparison to some other Member States. It could also show that the overall Dutch dairy sector stands to benefit from the abolition of the milk quota if the price effect is less than the volume effect.
Oudendag et al. (2014) argued that some small farmers are forced to exit the market when the quota is abolished. Their argument for this is that there are potential significant reductions in the profitability of milk production, due to lower milk prices. The result is that there will be an increase in the number of dairy cows per farm for the more efficient farmers
10 (Oudendag et al., 2014). My view on this by analyzing the studies of Chantreuil et al. (2008); Helming & van Berkum (2008); and Lips et al. (2005) is that it is highly likely that the Dutch farmers capture for a large extent on this production. A fact that can substantiates my view is that the type of quota that the Netherlands uses, a marketable quota, will most likely have resulted in more efficient production than the countries who use another type of quota (Wieck et al., 2007). To further support my statement, Samson et al. (2012) argued that the Dutch dairy sector is among the few where the quotas are still binding. The downside of this binding quota is that it influences the behavior of farmers since they cannot choose for optimal production levels themselves (Samson et al., 2012).
Profitable alternative use for the production inputs are relatively low in the Netherlands, and therefore the price of a quota relatively high (Oskam et al., 1992). The sales and lease prices in the Netherlands are therefore two to three times higher than the prices in the United Kingdom (Oskam et al., 1992). Without these quota prices, the Netherlands would produce at lower marginal cost which would lead to relative higher production (Oudendag et al., 2014). The decrease in quota costs may also outweigh milk price reduction for some small farmers, so that they can continue their production (Oudendag et al., 2014).
A counterargument that the liberalization of the production will likely result in higher production in the Netherlands is that the exit schemes had less influence in the Netherlands (Oskam et al., 1992). Exit schemes could contribute to faster structural development
because less efficient farmers who had been allocated a quota could choose during the first year of the regulation to exit the market and in return get a compensation (Oskam et al., 1992; Wieck, 2007). Looking at the quota price in the Netherlands relative to other Member States, I think that the effect that the exit schemes should have, is partially offset by the
marketable quota for the Dutch dairy sector (Oskam et al., 1992).
According to the population projection of the Eurostat (2014) and United Nations (2014) databases, the population in Europe is expanding. I state that the expanding population will highly likely result in higher demand for milk products. Therefore the increase in demand will partially compensate for decreasing milk prices (Bouamra-Mechemache et al., 2008a). According to Samson et al. (2012) the overall effect for the Dutch dairy sector, taking into account the expected increase in demand for milk, might lead to a 8 % decrease in milk price.
In summary, I think that it is highly likely that the production in the Netherlands will expand more relative to the overall European production expansion. One reason for this is that the quota is, in contrast to other Member State, still binding in the Netherlands (Samson et al., 2012). This could state that the Netherlands produce relatively more efficient. Thereby would, the decrease in profitability of milk production due to decreasing milk prices, force some small farmers out of the market. Because the Netherlands face relative more efficient producers, it could absorb the production of these small farmers. I also think that the effect of decreasing milk prices would partially be compensated by an increase in demand.
11
3. Method and Data
This chapter introduces the method used to answer the research question: “What will be the effect on the production of milk in the Netherlands when the milk quota will be abolished in 2015?” All in all, two regression analysis are carried out.
The first regression tries to explain the difference in future milk output without any regulation restrictions. I use the coefficients of the independent variables along with their projections to make future estimates of the milk output. There are several scenarios analyzed to see what the impact is for the milk output. Every scenario is examined yearly, namely from 2015 till 2019. The calculated projection of the milk output is also used in the second regression model as an explanatory variable to explain the milk prices.
The second regression tries to explain the development of the milk price based on the coefficients of the independent variables combined with their future projected values. Again, several scenarios emerged from the calculations. To see what the impact is on the milk price, the estimates of each scenario are examined yearly from 2015 till 2019.
The data used in the empirical model are obtained from different sources. I use mostly monthly data from 2001 up till 2014. For some variables the data couldn’t be obtained over the whole period. Other variables need linear interpolation to convert to monthly data. Most of the projections are obtained at the database self. For the remainder of the variables I consult the literature and used logical reasoning to make future estimates.
The models I intend to build at first deviates from the final models. This is due to the fact that several variables do not help to explain the dependent variables. Another reason is that there could exist multicollinearity between variables. Or the variables are hard to obtain like the marginal costs and the quota rent. Therefore several variables are omitted in the final models.
In the next section the explanatory variables that may have an impact on the production or the price are defined. A distinction is made between demand related and supply related variables. In the last sections the hypothesis for the empirical models are explained.
3.1 Variables and data sources Dependent Variable
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑒𝑒𝑡𝑡= 𝛽𝛽1+ 𝛽𝛽2∙ 𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝐿𝐿𝑡𝑡+ 𝛽𝛽3∙ 𝑀𝑀𝑄𝑄𝑃𝑃𝑄𝑄𝐿𝐿𝑡𝑡+ 𝛽𝛽4∙ 𝑀𝑀𝑄𝑄𝑃𝑃𝐸𝐸𝐸𝐸27𝑡𝑡+ 𝛽𝛽5
∙ 𝑀𝑀𝑄𝑄𝑃𝑃𝑅𝑅𝐸𝐸𝑡𝑡+ 𝛽𝛽6𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑄𝑄𝐿𝐿� + 𝑒𝑒𝑀𝑀𝑀𝑀𝑄𝑄𝑀𝑀 𝑄𝑄𝑒𝑒𝑀𝑀𝑚𝑚𝑡𝑡 𝑡𝑡
Milk Price in the Netherlands, MILK_PRICE
The monthly milk prices for the Netherlands are calculated by DG Agri and drawn from the DairyCo.org.uk database (DairyCo.org.uk, 2014c). The data used contains observations over the period from 2001 until April 2014. And the average milk price is given in euro’s per 100 kilograms.
Independent Variables
3.1.1 Demand-related variables
12 The data of the total quota allocated to the Netherlands is extracted from the website
ZuivelHistorieNederland.nl (ZuivelHistorieNederland, 2014). It contains yearly data from 2001 until 2011. To work with monthly data in the analysis, I use linear interpolations. The quota is measured in tons.
The allocated quota is used in the regression model because it is a constraint of the milk production, and would therefore likely adjust the milk price (Bouamra-Mechemache et al., 2008b; Oskam et al., 1992). After the quota is reached it is more costly for producers to supply another amount of milk due to the superlevy (Oskam et al., 1992). Hence, higher milk prices.
The future projection for the quota is based on the conclusion of Bouamra-Mechemache et al. (2008b) and Lips et al. (2005). Bouamra-Mechemache et al. (2008b) conclude that the market effects of quota abolishing in 2015 are comparable to a European wide 2 % gradual increase starting in 2009. Lips et al. (2005) concludes that several countries have incentives to expand production more than other countries. They argue that the milk production in Denmark, Ireland, Italy, Luxembourg, the Netherlands and Spain increases and in Germany, Greece, Portugal, and Sweden decreases. Therefore I state that the milk quota projections for the Netherlands after the last observation (January 2012) increase by 2% every year. Note that there will be no maximum quota anymore after 2015. The projections of the milk quota after 2015 can be interpreted as a “development curve of expansion”.
Population in the Netherlands, PopNLt
The monthly population per region is an indicator for the demand of milk. At first, I used every country of the European Union and Russia in my analyses. This was done because the Dutch dairy sector is an exporter to other European countries and Russia (CBS, 2014). But the correlations between these populations are too high. Therefore I only use the
populations of the EU-27, the Netherlands, and Russia in my analysis. The Russian
Federation is included because the Dutch dairy sector exports a great share of their milk to Russia (CBS, 2014). After adding the Russian Federation into the analyses the milk price for the Netherlands is better explained.
The data of the population and the future projections of the population per Member State are extracted from the Eurostat (2014) database. Only the future projections of the Russian Federation are extracted from the United Nations (2014) database. To convert the yearly data to monthly data, again I use linear interpolation.
3.1.2 Supply-related variables
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑄𝑄𝐿𝐿𝑡𝑡 = 𝛽𝛽1+ 𝛽𝛽2∙ 𝐿𝐿𝑄𝑄𝐿𝐿𝐿𝐿𝑀𝑀𝑄𝑄𝐿𝐿𝑄𝑄𝐿𝐿𝑡𝑡+ 𝛽𝛽3∙ 𝐶𝐶𝑄𝑄𝐶𝐶𝑄𝑄𝐿𝐿𝑡𝑡−15+ 𝛽𝛽4∙ 𝐶𝐶𝑄𝑄𝐶𝐶𝐶𝐶𝑄𝑄𝑀𝑀𝑀𝑀𝐿𝐿𝑡𝑡−15+ 𝛽𝛽5
∙ 𝐶𝐶𝑄𝑄𝐶𝐶𝐸𝐸𝐸𝐸𝑡𝑡−15+ 𝛽𝛽6∙ 𝐿𝐿𝑄𝑄𝐿𝐿𝑄𝑄𝑀𝑀𝐶𝐶𝑄𝑄𝐿𝐿𝑄𝑄𝐿𝐿𝑡𝑡−3+ 𝑒𝑒𝑀𝑀𝑀𝑀𝑄𝑄𝑀𝑀 𝑄𝑄𝑒𝑒𝑀𝑀𝑚𝑚𝑡𝑡
Total Milk Production in the Netherlands, MilkNLt
In my first model the milk production for the Dutch dairy sector is the dependent variable. The variables that explain the milk production are: land costs in the Netherlands; milk producing cows in the Netherlands; milk producing cows in the world; milk producing cows in the European Union; and labor costs in the European Union. Note that the milk
13 The total milk production for the Netherlands is assembled from the DairyCo
(DairyCo.org.uk, 2014a) database and is given in tons. The future projections of the milk production, is estimated by the first regression model.
At first I also wanted to include “feed prices” in the model but they are hardly to obtain for the Netherlands. Note that feed prices outside the Netherlands could neither be used because they are non-comparable. I think that it is not a problem to omit the feed prices since they are partly embedded in the land costs, every land is used to produce nutrition for the cows. Therefore I conclude not to use any feed prices in my model.
Land Costs in the Netherlands, LandCostst
Boots et al. (1996) state that the milk price is partly determined by the labor-, land-, building-, and machinery costs. In this model, only the labor- and land costs are taken into consideration, because it is hard to obtain data for the building- and machinery costs. The costs in euro’s to obtain a hectare of land in the Netherlands are extracted from the Eurostat (2014b) database. I made use of linear interpolation to get monthly data.
The land prices in the future can increase or decrease. Therefore three scenarios are made for the land prices. Scenario one assumes that the price of land increase at the same pace of the last three years. This is done by the growth function in Excel. How good the forecasted values fit the original data is shown in graph 3.1a of the appendix. The forecasted values of the cost of land till 2019 are shown in graph 3.1b. Scenario two assumes that the price stays constant after the last observation, at 47.051 €/Ha. And scenario three assumes that the land price face an instant decrease to the lowest value observed on January 2006, namely 30.235 €/Ha.
At first the idea was to use the land costs for every EU-15 country, to include some information of the relative advantage of lower marginal costs for one country compared to the other. But the land costs for different countries have a strong correlation. Therefore I left out the land prices for other EU-15 countries. Note that this doesn’t need to have negative consequences for my model because my models tries to explain the Dutch milk price and Dutch milk production.
Labor Costs in the EU-17, LaborCostst-3
To embed the labor costs in my model I used the monthly labor costs values for the EU-17 obtained via the database of the European Central Bank (2014). A lag of 3 months for the labor costs explains the Dutch milk price and Dutch milk production better, compared with no delay. The reason for this may be contract rigidities. So that employees must comply to contracts and cannot quit their job without a notice period of 3 months.
Because it is hard to get future projections, the assumption will be made that labor costs stay constant after January 2014. I don’t think that this has negative consequences for my model because the significance of the labor costs is low, which could be the case, because most Dutch farmers don’t have additional employees (Samson et al., 2012).
Number of Cows for Several Regions, CowWorldt-15, CowNLt-15, CowEUt-15
In my model I use the total number of cows in the entire world, the total number of cows in the Netherlands, and the total number of cows in the European Union. This data is extracted from the DairyCo database (DairyCo.org.uk, 2014b), and converted to monthly data.
Because it takes 15 months before cows can produce milk, the data of total cows per region all include a 15th lag to exclude non-producing cows. So by introducing a 15th lag only
14 valuable information will remain. Hence, the model shows a great improvement of the coefficient of determination.
There is a slight increase of the number of cows in the Netherlands starting from 2011. This increase is probably due to the Dutch farmers expansion strategy, to exploit economies of scale (Samson et al., 2012). Oudendag et al., (2014) also argued the expansion of the number of cows as a result of the increase in milk quotas. Therefore, I conclude that the future number of cows in the Netherlands will slightly increase starting from 2011, combined with a lower rate of increase every month. This is done by using the growth rate function in Excel over the period from 2011 till 2013 combined with a growth rate decline starting in April 2013. The maximum number of cows is then reached in the half of 2015 and remains stable at 1,653,112 cows for the Netherlands. This is a 3.5% increase since January 2013. Graph 3.2a shows the projection of the number of cows in the Netherlands and how the original data fit the forecasts.
For the European Union 15 Member States it is assumed that the future number of cows will decline with the same pace as can be seen between 2008 and 2011 (graph 3.2d). This is done because over the whole period the number of cows is declining. By using the growth function in Excel over the period 2008 till 2011, the number of cows in the EU is 16.238.794 at the beginning of 2019. Graph 3.2b shows how the forecasted values fit the original data. Graph 3.2c shows the forecasted number of cows for the EU-15.
The number of cows in the world has increased until 2009 (graph 3.3a). From 2009 till 2011 it stayed more or less constant. Therefore I made two scenarios. One scenario
calculates the impact when the number of cows stays constant, at 260.000.000 cows, after 2011. And the second scenario where the number of cows in the entire world increases based on the pace of the first period (from 2001 until 2008). Graph 3.3a shows how the predicted values fit the real data. Graph 3.3b shows the future predictions of the total amount of cows.
3.2 Milk output changes due to the changing production inputs
With the first regression I analyze how the milk output reacts due to changes in the explanatory variables (shown in the regression model below). The method that I use to perform the regression is a time series analyses. After the regression I made a future outlook of the milk production in the Netherlands based on the projections of the explanatory
variables. The projections of the explanatory variables are made by logical reasoning and the literature as explained above. The hypothesis is defined as follows:
H0: Milk production stay constant after the abolishment H1: Milk production increase after the abolishment Regression model 1:
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑄𝑄𝐿𝐿𝑡𝑡 = 𝛽𝛽1+ 𝛽𝛽2∙ 𝐿𝐿𝑄𝑄𝐿𝐿𝐿𝐿𝑀𝑀𝑄𝑄𝐿𝐿𝑄𝑄𝐿𝐿𝑡𝑡+ 𝛽𝛽3∙ 𝐶𝐶𝑄𝑄𝐶𝐶𝑄𝑄𝐿𝐿𝑡𝑡−15+ 𝛽𝛽4∙ 𝐶𝐶𝑄𝑄𝐶𝐶𝐶𝐶𝑄𝑄𝑀𝑀𝑀𝑀𝐿𝐿𝑡𝑡−15+ 𝛽𝛽5
∙ 𝐶𝐶𝑄𝑄𝐶𝐶𝐸𝐸𝐸𝐸𝑡𝑡−15+ 𝛽𝛽6∙ 𝐿𝐿𝑄𝑄𝐿𝐿𝑄𝑄𝑀𝑀𝐶𝐶𝑄𝑄𝐿𝐿𝑄𝑄𝐿𝐿𝑡𝑡−3+ 𝑒𝑒𝑀𝑀𝑀𝑀𝑄𝑄𝑀𝑀 𝑄𝑄𝑒𝑒𝑀𝑀𝑚𝑚𝑡𝑡
15 In the second regression I analyze how the milk price is determined based on the
explanatory variables (shown in the regression model below). This is also done by a time series analyses. After I define the importance of each variable, I made an estimate of the future milk price based on the projections of the explanatory variables. The projections of the explanatory variables are made by logical reasoning and the literature as explained above. The hypothesis is defined as follows:
H0: Milk price has not changed due to increase in quota H1: Milk price has changed due to increase in quota Regression model 2:
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑒𝑒𝑡𝑡= 𝛽𝛽1+ 𝛽𝛽2∙ 𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝐿𝐿𝑡𝑡+ 𝛽𝛽3∙ 𝑀𝑀𝑄𝑄𝑃𝑃𝑄𝑄𝐿𝐿𝑡𝑡+ 𝛽𝛽4∙ 𝑀𝑀𝑄𝑄𝑃𝑃𝐸𝐸𝐸𝐸27𝑡𝑡+ 𝛽𝛽5
∙ 𝑀𝑀𝑄𝑄𝑃𝑃𝑅𝑅𝐸𝐸𝑡𝑡+ 𝛽𝛽6𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑄𝑄𝐿𝐿� + 𝑒𝑒𝑀𝑀𝑀𝑀𝑄𝑄𝑀𝑀 𝑄𝑄𝑒𝑒𝑀𝑀𝑚𝑚𝑡𝑡
3.4 Possible effects of the quota’s abolishment after 2015
At last, I examine the production and price effect to investigate the effects for the future production in the Netherlands. If the price effect dominates the production effect, than it will have a negative consequence for the Dutch dairy sector. The hypothesis is defined as follows:
H0: Future production will have a negative impact due to the abolishing of the quota H1: Future production will have a positive impact due to the abolishing of the quota
16
4. Findings
In this section the findings obtained are discussed. In the next paragraph the two models will be discussed. The second paragraph answers the hypothesis: “how the milk output
changes”. The second hypothesis: “how the milk price changes” is answered in the third paragraph. In the last section the price effect and the production effects will be examined to give an answer to the research question.
4.1 Results of the models
The first model investigates the importance of the explanatory variables on the milk production in the Netherlands. The results of the regression (Table 4.1a) shows that the R-square of the regression is 0.9873. This explains how well the data fits the statistical model and how well the independent variables explain the dependent variable. In other words, about 98% of the dependent variable is explained by the explanatory variables.
The observed period contains 161 observations, but because of missing data for several variables for several periods the number of observations taking into account in the
regression declines. Also the lags within the model lowers the numbers of observations that can be used. I end up with a total of 94 observations to predict the milk production.
At first it could be contra intuitive that increasing land prices result in more milk production. But when there is more demand for milk, more milk would be produced and more land is needed. In other words, land is then more profitable in the hands of the farmers, so farmers ask a higher price for the land. Another explanation could be omitted variable bias.
The coefficient of the total number of cows in the Netherlands is positive. This means that there will be more milk produced in the Netherlands when there are more cows in the Netherlands. On the other hand, milk production for the Netherlands will decline when other regions increase their number of cows. The Dutch milk production will then be absorbed by regions outside the Netherlands. This development can be interpreted by the minus sign of the total number of cows in the World and European Union.
The coefficient of the labor costs show that when labor costs rises, milk production will rise. This is probably caused by the relation between labor costs and the state of the
economy. When the economy is growing, more milk will be produced and employees will get better paid. The significance of the labor costs is low, which could be the case, because most Dutch farmers don’t have additional employees (Samson et al., 2012).
17
Table 4.1a Results of the regressions
MilkNLt MilkPricet (1) (2) LandCostst 36,405060*** (1,404116) CowNLt-15 1,0043940*** (0,338256) CowWorldt-15 -0,0178635*** (0,004676) CowEUt-15 -0,4433000*** (0,062826) LaborCostst-3 4024,4160 (4241,480) QuotaNLt 0,4587950*** (0,010031) PopNLt -0,0000680*** (0,000016) PopEU27t 0,0000020*** (0,000000) PopRUt 0,0000026** (0,000001) MılkNL� t -0,0000229*** (0,000007) Intercept 20900000 -511,6732 (1798657) (397,0423) No. Obs. 94 94 R-Squared 0,9873 0,6046 F-Statistic 0,0000 0,0000
Notes: The data are described in the Method and Data section. The second and third column contains the coefficients of the explanatory variables, with standard errors in the parenthesis. The final three rows report the number of observations, the R-Squared, and the F-statistic testing the hypothesis that all the coefficients in the regression are zero. Significance at the 1%, 5%, and 10% levels are denoted respectively by ***, **, and *.
The second model investigates the importance of the explanatory variables on the Dutch milk price. The results of the regression (Table 4.1a) shows that the coefficient of
determination is 0.6046. This means that 60% of the dependent variable is explained by the explanatory variables. The model uses also 94 observation for the same reasons as the model above.
The coefficient of the milk quota is contra intuitive. According to Snyder et al. (2012) the price of a good will increase when the quota declines. This is not the case in my model. Maybe the reason for this is that the milk production in the Netherlands is higher than the quota permits (DairyCo.org.uk 2014a; ZuivelHistorieNederland, 2014).
According to the model, an increase in population in the European Union and the Russian Federations results in an increase of the milk price. This is consistent with the theory of Snyder et al. (2012). The model state further that the increase in population in the
Netherlands results in lower milk prices. I think that the reason for this reaction is that there are high correlations between the three regions as shown in table 4.1b. So that in the same period an increase in demand in the Netherlands will be offset by a larger decrease in
18 small, as compared to the other regions. The negative coefficient would not have an impact on my model because the increase and decline in population for different regions are approximately steady as shown in graph 4.1.
Milk output is based on the first regression. The second model state that an increase of the milk output will result in lower milk prices. This is also observed by the studies of Bouamra-Mechemache et al. (2008b); Chantreuil et al. (2008); Helming & van Berkum (2008); Lips et al. (2005); and Samson et al. (2012). What the impact of the increasing milk production is, will be analyzed in paragraph 4.3.
Table 4.1b Correlation between different populations
PopNLt PopRUt PopEU27t
PopNLt 1.0000
PopRUt -0.8005 1.0000
PopEU27t 0.9771 -0.8888 1.0000
4.2 How the milk output change
By analyzing the results of the regression and performing scenario analysis, I will give an answer to the first hypothesis:
H0: Milk production stay constant after the abolishment H1: Milk production increase after the abolishment
The six scenarios that are analyzed are shown in the diagram below:
Diagram 4. 2a Six scenarios for the milk production
Labour Costs for the EU-17 Number of Cows in the EU-15
Number of Cows in the NL Number of Cows in the World
Prices of Land
Start
Instant decline to 30.235 €/Ha Constant Increasing Decreasing Constant Scenario 1 Increasing Increasing Decreasing Constant Scenario 2 Constant at 47.051 €/Ha Constant Increasing Decreasing Constant Scneario 3 Increasing Increasing Decreasing Constant Scenario 4 Increasing Constant Increasing Decreasing Constant Scenario 5 Increasing Increasing Decreasing Constant Scenario 619 The percentage change in milk production with respect to the base year of 2012 is shown in the diagram below.
According to the results of the calculations for each scenario, the milk production in the Netherlands will rise if the total number of world cows stays constant. The results state also that if the total number of cows in the world would rise the output in the Netherlands would be suppressed.
The regression output in figure 4.1a state that the increase in land costs will increase milk production, this is also shown in the results shown in diagram 4.2b. I think that scenario 1 and 2 are less likely than the other scenarios. This is because land prices will rise after the abolishment of the quota due to increase in demand for land. Note that the quota is still binding in the Netherlands and farmers want to produce more to reach efficient levels (Samson et al., 2012). However, I still use the two scenarios in the next model to see what the impact is.
Scenario 1, 3, 5 and 6 gives an expansion in milk production. Therefore I conclude that it is highly likely that there will be more milk production in the Netherlands after the quota abolishment. The zero hypothesis will thus be rejected.
Diagram 4. 2b Results of the milk production for the six scenarios
4.3 How the milk price change
By analyzing the results of the regression and perform the scenarios that are given in paragraph 4.2 on the milk price, I will give an answer to the second hypothesis: H0: Milk price has not changed due to increase in quota
H1: Milk price has changed due to increase in quota
2013 2016 2019 2018 2014 2017 2015 2012 Scenario 1 -3.10% -2.17% -1.31% -0.64% 0% 0.64% 1.28% Scneario 2 -5.67% -5.48% -5.38% -5.49% -5.64% -5.8% -5.98% Scenario 3 2.14% 3.08% 3.94% 4.6% 5.25% 5.89% 6.52% Scenario 4 -0.42% -0.23% -0.14% -0.25% -0.39% -0.55% -0.74% Scenario 5 7,93% 11.46% 15.25% 19.21% 23.57% 28.40% 33.75% Scenario 6 5.37% 8.15% 11.18% 14.36% 17.93% 21.95% 26.49%
20
Diagram 4.3 Difference between the calculation based on the regression output and the reality
Graph 4.3 Milk price scenarios
The output of the different scenarios are shown in graph 4.3. According to the scenarios the milk price will increase heavily in scenarios 1,2,3 and 4. This is largely a result of the
absorbed milk production outside the European Union. According to the model the price would therefore rise, which isn’t logical. This is because my model doesn’t account for the price development if the milk will be imported outside the European Union. But what is embedded in scenarios 1, 3, 5 and 6, is that an increase in milk production in the European Union will result in a decline in the milk price. The effect is less observable in scenarios 1 and 3, because the production outside the European Union predominates.
I conclude that scenarios 5 and 6 are the most logical ones. Despite the fact that demand will increase in the European Union, milk prices will decline. Scenario 5 and 6 do not embed the price development if the milk will be imported outside the European Union, so the price would decline even further (Snyder et al., 2014). Therefore I will reject the zero hypothesis and state that the milk price would decline as a response to higher milk production.
2010 2013 2011 2014 2012 Start Model € 24.27 € 29.31 € 33.99 € 22.76 € 32.73 Reality € 27.01 € 31.12 € 31.25 € 30.53 € 36.65 € 0,00 € 20,00 € 40,00 € 60,00 € 80,00 € 100,00 € 120,00 Scenario 1 Scenario 2 scenario 3 Scenario 4 Scenario 5 Scenario 6
21 4.4 How would future production develop
Based on the outcomes of the two models and the literature review I give a answer to the third hypothesis:
H0: Future production will have a negative impact due to the abolishing of the quota H1: Future production will have a positive impact due to the abolishing of the quota According to the outcome of the first hypothesis, milk production will increase in the
Netherlands. This is also predicted by the studies of Helming & van Berkum (2008); Lips et al. (2005); and Samson et al. (2012). I think that, based on the high quota price and the binding quota for the Dutch dairy sector, it is highly likely that the Netherlands will expand
production more than the average production expansion in the European Union (Samson et al., 2012). Helming & van Berkum (2008) and Lips et al. (2005) also came to the conclusion that the production in the Netherlands will increase more than in most other countries. Therefore I conclude that the increase in production will highly likely have a greater impact for Dutch farmers, than the decrease in price would have. Hence, I reject the zero
hypothesis, and conclude that the abolishing of the quota will have a positive impact on future production.
The statements given by the models could be true. It could also be that the models are incomplete and have left out important causal variables to explain the dependent variable, this is called omitted variable biases, for instance policy regulation. If farmers for some reason are restricted by fertilizer use, they cannot produce as much milk as the model stated. According to Parton (1992), government intervention in the agricultural sector grossly distorts the market system, so that milk prices become largely policy determined. So the effect after 2015 could be a lot different, than sketched here.
22
5. Conclusive Remarks
First, I conveyed a literature review about the implementation and mechanism of the quota system in the Netherlands, and the differences between other Member States’ quota systems. Thereafter I performed an empirical research. The hardest part was to determine which variables to use, and to obtain monthly data for these variables. The literature review has helped me with this, so I could convey my own empirical research. My empirical
research is about the establishment of the Dutch milk production and the Dutch milk price. To analyze these parameters I made use of a time series analyses. In Stata I was able to see the importance of every variable and how good they explain the parameter. Based on the coefficients and projections of each explanatory variable, I could state something about the future development of both: the milk production; and the milk price.
It was easy to construct a model for the supply side. But when supply and demand were combined to investigate how the price is affected, it was more difficult to determine suitable variables. In the end I think that I have succeeded to make two compact models. The only con of the second model is that import by the dairy processors cannot be explained by the model. This is because my model doesn’t account for the price development if the milk will be imported outside the European Union. Hence, it give nonrealistic outcomes when total number of cows in the world increases. Therefore I combine in the last section of the “Obtained Findings”, the literature review with my own empirical research to give a final conclusion to the research question: “What will be the effect on the production of milk in the Netherlands when the milk quota will be abolished in 2015?”.
The conclusion of my first model is that milk production will increase in the Netherlands, based on the projections of the explanatory variables. According to the second model, excluding the total number of cows in the world, the increase in production will result in lower prices. My empirical research does not include information about the rest of the European Union. Hence, I used the literature to state something about the change in milk production in the rest of Europe. According to the studies of Chantreuil et al. (2008); Helming & van Berkum (2008); and Lips et al. (2005) is that the Dutch milk production will likely increase more than the average increase in production in the European Union. This could be a byproduct of the almost free quota market in the Netherlands, which is active since the 1990s (Oskam et al., 1992; Wieck et al., 2007). This has led to faster structural development so that most Dutch farmers can produce relatively efficient (Oskam et al., 1992). Their marginal costs are lower and they are willing to pay more for a quota than farmers in other countries would pay (Oskam et al., 1992; Wieck et al., 2007). Also the quota in the Netherlands is, in contrast to most other countries, still binding (Samson et al.,
2012).Therefore it is highly likely that the Dutch dairy sector absorb more production relative to other countries.
According to the population projection of the Eurostat (2014) and United Nations (2014) databases, the population in Europe is expanding. Therefore the demand for milk increases. So the price decline will be partly offset by increase in demand for milk
(Bouamra-Mechemache et al., 2008a).
Hence I conclude, that based on the literature and my empirical research, that there is a positive impact for the future Dutch milk production. This is because the Dutch dairy sector will likely absorb more of the production of the inefficient farmers than most countries. And the population increase will partly offset the price decline. Hence, the Dutch dairy sector ends up with a production effect that dominates the price effect.
23 It could be that the results of my model are distorted. Because my models does not include any government interventions active for the period observed, despite the fact that prices could be policy determined (Parton, 1992). Another possible flaw is that there are other important causal factors that aren’t observed in the model, the so called omitted variable biases. Therefore the observed phenomenon by the models could be biased.
For further research of the impact of the milk quota abolishment, scenarios can be made with other explanatory variables. I left out the building and machinery costs, because I could not obtain the data. According to Boots et al. (1997) the building and machinery costs have a significant impact on the marginal costs farmers face. Another improvement would be if the analyses embed the relative incentive that farmers have in different Member States to increase production. This could be done by comparing monthly data of the quota rents as a percentage of the marginal costs or prices between the different Member States. According to Parton (1992), milk price is largely policy determined. Therefore it would be a great improvement to map all policies and regulations active in the observed period. The last improvement that I consider is including the structural development into the model. By creating a relation between total production per region and the number of cows per region.
24
Reference List
Babcock, B. A., & Foster, W. E. (1992). Economic Rents under Supply Controls with Marketable Quota. American Journal of Agricultural Economics, 74(3), 630-637. Boots, M., Oude Lansink, A., & Peerlings, J. (1997). Efficiency loss due to distortions in Dutch
milk quota trade. European Review of Agricultural Economics, 24(1), 31-46. Bouamra-Mechemache, Z., & Réquillart, V. (2007). The dairy industry in a fast-growing
Euopean Union: policies and strategies. INRA Sciences Sociales, 2-3, 1-5.
Bouamra-Mechemache, Z., Chavas, J., Cox, T. L., & Réquillart, V. (2008a, August). Removing
EU milk quotas, soft landing versus hard landing. Paper presented at the 12th
Congress of the European Association of Agricultural Economists, Ghent, BE. Bouamra-Mechemache, Z., Jongeneel, R., & Réquillart, V. (2008b). Impact of a gradual
increase in milk quotas on the EU dairy sector. European Review of Agricultural
Economics, 35(4), 461-491.
CBS. (2014). Zuivelgroothandel zet 12 miljard euro om. Retrieved from http://www.cbs.nl/nl- NL/menu/themas/handel-horeca/publicaties/artikelen/archief/2014/2014-4059-wm.htm Accessed June 2014. NL.
Colman, D. (2000). Inefficiencies in the UK milk quota system. Food Policy, 25(1), 1-16. Chantreuil, F., Donnellan, T., van Leeuwen, M., Salamon, P., Tabeau, A., & Bartova, L. (2008,
August). EU Dairy Quota Reform – AGMEMOD Scenario Analysis. Paper presented at the 12th Congress of the European Association of Agricultural Economists, Ghent, BE. DairyCo.org.uk (2014a). World Milk Production, 1961-2012. Retrieved from
http://www.dairyco.org.uk/resources-library/market-information/supply-production/world-milk-production/#.Uvou_fkyDQo. Accessed June 2014 DairyCo.org.uk (2014b). World Cow Numbers, 2000-2013. Retrieved from
http://www.dairyco.org.uk/resources-library/market-information/farming-data/world-cow-numbers/. Accessed June 2014.
DairyCo.org.uk (2014d). DairyCo Database, 2001-2014. Retrieved from http://www.dairyco.org.uk/resources-library/. Accessed June 2014. DairyCo.org.uk (2014c). EU Milk Prices - DG Agri, 2001-2014. Retrieved from
http://www.dairyco.org.uk/resources-library/market-information/milk-prices-contracts/eu-milk-prices-dg-agri/. Accessed November 2014.
25 European Central Bank (2014). Labour Costs in EU-17. Retrieved from
http://sdw.ecb.int/quickview.do?SERIES_KEY=119.ESA.Q.I6.Y.1000.UNLACO.0000.TTT
T.D.U.R&start=01-01-2001&end=01-09-2014&trans=N&submitOptions.x=0&submitOptions.y=0. Accessed June 2014. European Commission. (2000). The common agricultural policy – 2000 Review. European
Commission : Directorate-General Agricultural Retrieved from
http://ec.europa.eu/agriculture/publi/review00/full_en.pdf Accessed June 2014. Luxembourg, LU.
European Commission. (2009). Economic Impact of the Abolition of the Milk Quota Regime:
Regional Analysis of the Milk Production in the EU. Retrieved from
http://ec.europa.eu/agriculture/analysis/external/milkquota/full_report_en.pdf Accessed June 2014. Seville, ES.
European Commission. (2014). CAP expenditure in the total EU expenditure, 1980-2011.
Brussels: and Rural Development. Retrieved from
http://ec.europa.eu/agriculture/cap-post-2013/graphs/graph1_en.pdf Accessed June 2014.
Eurostat. (2014). Agricultural Database for the European Union, 2001-2014. Retrieved from http://epp.eurostat.ec.europa.eu/portal/page/portal/statistics/search_database. Accessed June 2014.
Eurostat. (2014b). Land prices and rents, 2001-2010. Retrieved from
http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=apri_ap_aland&lang=en Accessed June 2014.
Helming, J. F. M., & Beerkum van, S. (2008, August). Effects of abolition of the EU milk quota
system for Dutch agriculture and environment. Paper presented at the 12th Congress
of the European Association of Agricultural Economists, Ghent, BE.
Hollander, A. (1993). Restricting Intra-Industry Quota Transfers in Agriculture: Who Gains, Who Loses?. The Canadian Journal of Economics, 26(4), 969-975.
Krugman, P.R., Obstfeld, M., & Melitz, M.J. (2012). International Economics – Theory &
Policy. Amsterdam: Pearson. pp. 222-248.
Lips, M., & Rieder, P. (2005). Aboiltion of Raw Milk Quota in the European Union: A CGE Analysis at the Member Country Level. Journal of Agricultural Economics, 56(1), 1-17.
26 Oskam, A. J., & Speijers D. P. (1992). Quota mobility and quota values – Influence on the
structural development of dairy farming. Food Policy, 17(1), 41-52.
Oudendag, D., Hoogendoorn, M., & Jongeneel, R. (2014). Agent-Based Modeling of Farming Behavior: A Case Study for Milk Quota Abolishment. Modern Advances in Applied
Intelligence, 8481, 11-20.
Parton, K. A. (1992). EC Diary Policy – an intergrated supply and policy analysis. Food Policy,
17(3), 187-200.
Samson, G. S., Gardebroek, C., & Jongeneel, R. (2012, June). The Cost Function Structure of
Dutch Dairy Farms: Effects of Quota Abolition and Price Volatility. Paper prepared for
the 126th Seminar of the European Association of Agricultural Economists, Capri, IT. Snyder, C. & Nicholson, W. (2012). Microeconomic Theory – Basic Principles and Extensions.
Amsterdam: Cencage. pp. 135-409.
United Nations. (2014). World Populations Prospects: The 2012 Revisions, 2000-2025. Retrieved from http://esa.un.org/wpp/unpp/panel_population.htm Accessed July 2014.
Wieck, C., & Heckelei, T. (2007). Determinants, differentiation, and development of short-term marginal costs in dairy production: an empirical analysis for selected regions of the EU. Agricultural Economics, 36(2), 203-220.
ZuivelHistorieNederland. (2014). Allocated Quotas to the Netherlands. 2001-2011. Retrieved from http://www.zuivelhistorienederland.nl/ Accessed at June 2014.
27
Appendix
Graph 3. 1a Forecast (increase) prediction of land costs compared to original data
Graph 3. 1b Forecast prediction of land costs till 2019
20.000,0 25.000,0 30.000,0 35.000,0 40.000,0 45.000,0 50.000,0 ja n-0 6 ap r-0 6 ju l-0 6 okt -0 6 ja n-0 7 ap r-0 7 ju l-0 7 okt -0 7 ja n-0 8 ap r-0 8 ju l-0 8 okt -0 8 ja n-0 9 ap r-0 9 ju l-0 9 okt -0 9 ja n-1 0 Original Forecast 0,0 20.000,0 40.000,0 60.000,0 80.000,0 100.000,0 120.000,0 140.000,0 160.000,0 ja n-0 6 se p-0 6 m ei -07 ja n-0 8 se p-0 8 m ei -09 ja n-1 0 se p-1 0 m ei -11 ja n-1 2 se p-1 2 m ei -13 ja n-1 4 se p-1 4 m ei -15 ja n-1 6 se p-1 6 m ei -17 ja n-1 8 se p-1 8 Original Forecast
28
Graph 3. 2a Forecast prediction of the number of cows in the Netherlands
Graph 3. 2b Forecast prediction of number of cows in the EU-15 compared to original data
1.490.000,0 1.510.000,0 1.530.000,0 1.550.000,0 1.570.000,0 1.590.000,0 1.610.000,0 1.630.000,0 1.650.000,0 1.670.000,0 ja n-1 1 ju l-1 1 ja n-1 2 ju l-1 2 ja n-1 3 ju l-1 3 ja n-1 4 ju l-1 4 ja n-1 5 ju l-1 5 ja n-1 6 ju l-1 6 ja n-1 7 ju l-1 7 ja n-1 8 ju l-1 8 ja n-1 9 ju l-1 9 ja n-2 0 Forecast Original 17.200.000,0 17.300.000,0 17.400.000,0 17.500.000,0 17.600.000,0 17.700.000,0 17.800.000,0 17.900.000,0 18.000.000,0 18.100.000,0 18.200.000,0 18.300.000,0 ja n-0 8 m rt -0 8 m ei -08 ju l-0 8 se p-0 8 no v-08 ja n-0 9 m rt -0 9 m ei -09 ju l-0 9 se p-0 9 no v-09 ja n-1 0 m rt -1 0 m ei -10 ju l-1 0 se p-1 0 no v-10 ja n-1 1 Original Forecast
29
Graph 3. 2c Forecast prediction of number of cows in the EU-15 till 2019
Graph 3. 2d Number of cows in the EU-15
Graph 3. 3a Forecast prediction of number of cows in world compared to original data
15.000.000,0 16.000.000,0 17.000.000,0 18.000.000,0 19.000.000,0 20.000.000,0 21.000.000,0 ja n-0 1 ja n-0 2 ja n-0 3 ja n-0 4 ja n-0 5 ja n-0 6 ja n-0 7 ja n-0 8 ja n-0 9 ja n-1 0 ja n-1 1 ja n-1 2 ja n-1 3 ja n-1 4 ja n-1 5 ja n-1 6 ja n-1 7 ja n-1 8 ja n-1 9 Original Forecast 16.000.000,0 17.000.000,0 18.000.000,0 19.000.000,0 20.000.000,0 21.000.000,0 ja n-0 1 ju l-0 1 ja n-0 2 ju l-0 2 ja n-0 3 ju l-0 3 ja n-0 4 ju l-0 4 ja n-0 5 ju l-0 5 ja n-0 6 ju l-0 6 ja n-0 7 ju l-0 7 ja n-0 8 ju l-0 8 ja n-0 9 ju l-0 9 ja n-1 0 ju l-1 0 ja n-1 1 ju l-1 1 ja n-1 2
Number of cows per country (source: FAO/dairy.co) Yearly
EU-15
210.000.000,0 220.000.000,0 230.000.000,0 240.000.000,0 250.000.000,0 260.000.000,0 270.000.000,0 ja n-0 1 ju l-0 1 ja n-0 2 ju l-0 2 ja n-0 3 ju l-0 3 ja n-0 4 ju l-0 4 ja n-0 5 ju l-0 5 ja n-0 6 ju l-0 6 ja n-0 7 ju l-0 7 ja n-0 8 ju l-0 8 ja n-0 9 Original Forecast30
Graph 3. 3b Forecast prediction of total number of cows in the world till 2019
Graph 3. 3c Number of cows in the world
150.000.000,0 170.000.000,0 190.000.000,0 210.000.000,0 230.000.000,0 250.000.000,0 270.000.000,0 290.000.000,0 310.000.000,0 330.000.000,0 ja n-0 1 fe b-0 2 m rt -0 3 ap r-0 4 m ei -05 jun-06 ju l-0 7 au g-0 8 se p-0 9 okt -1 0 no v-11 de c-1 2 ja n-1 4 fe b-1 5 m rt -1 6 ap r-1 7 m ei -18 Original Forecast Increase Forecast Constant 200.000.000,0 210.000.000,0 220.000.000,0 230.000.000,0 240.000.000,0 250.000.000,0 260.000.000,0 270.000.000,0 ja n-0 1 ju l-0 1 ja n-0 2 ju l-0 2 ja n-0 3 ju l-0 3 ja n-0 4 ju l-0 4 ja n-0 5 ju l-0 5 ja n-0 6 ju l-0 6 ja n-0 7 ju l-0 7 ja n-0 8 ju l-0 8 ja n-0 9 ju l-0 9 ja n-1 0 ju l-1 0 ja n-1 1
Number of cows per country (source: FAO/dairy.co) Yearly
World
31
Graph 4. 1 Populations of different regions
15.400.000 15.600.000 15.800.000 16.000.000 16.200.000 16.400.000 16.600.000 16.800.000 17.000.000 17.200.000 17.400.000 ja n-0 1 fe b-0 2 m rt -0 3 ap r-0 4 m ei -05 jun-06 ju l-0 7 au g-0 8 se p-0 9 okt -1 0 no v-11 de c-1 2 ja n-1 4 fe b-1 5 m rt -1 6 ap r-1 7 m ei -18