Phasing out the EU milk quota: its effect on productivity in organic relative to conventional EU dairy farming

UNIVERSITY OF AMSTERDAM

Phasing out the EU milk quota: its effect on

productivity in organic relative to conventional EU

dairy farming

Bachelor thesis Economics

Ellen van ‘t Klooster Student number: 10207309 With the help and supervision of:

Dhr. J. Sun Msc

06/28/2016

Abstract

In 2004, the EU started loosening its milk quota. To analyze the effect of the relaxing of the milk quota on competitiveness of organic relative to conventional dairy farms, I analyzed and compared data on EU dairy farms during the period 2004-2011. I determined total

productivity and used a Cobb-Douglas production function with a time trend to determine changes in total factor productivity (TFP). The results show the same trends in total

productivity and TFP for organic and conventional dairy farms in 2004-2011, suggesting that the phasing out of the milk quota does not affect competitiveness of organic relative to conventional dairy farms. TFP changes are negative for both types, despite an increase in milk yields. This can be explained by a strong input growth in 2004-2009.

2

Statement of originality

This document is written by Student Ellen Lydia van ‘t Klooster who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

3

List of contents

2.1 Measuring productivity in dairy farming ... 6

2.2 Comparisons of organic and conventional dairy farming ... 7

2.3 Phasing out the EU milk quota ... 8

3. Methods... 10

3.1 Measuring total productivity ... 10

3.2 Measuring Total Factor Productivity ... 10

4. Data ... 12

4.1 Dataset Farm Accountancy Data Network ... 12

4.2 Milk prices ... 13

4.3 Descriptive statistics ... 13

5. Results ... 15

5.1 Total productivity ... 15

5.2 Total factor productivity ... 16

6. Discussion ... 19

7. Conclusion ... 21

References ... 22

4

1. Introduction

In 2013, the EU produced more than 210 million tons of milk, representing 33% of world cow milk production. This means the EU is the biggest producer of milk in the world.

(FAOSTAT, 2013) An important development in EU dairy production is the loosening of the milk quota since 2004 and its abolishment in 2015. Another current development is the stimulation and growth of the organic dairy industry. Breustedt, Latacz-Lohmann and Tiedemann (2011) state that abolishment of the milk quota reduces competitiveness of organic dairy farming.

Organic dairy farming began in the 1990s. (Pierce J., 2015) Among the current cow’s milk producers, an increasing percentage is converting from conventional methods to organic methods. (Breustedt et al., 2011) Figure 1 demonstrates that the quantity of organic milk produced in the EU has increased significantly since 2007. In 2014, organic dairy cow’s milk production was 3.8 million metric ton, which represented more than 2.6% of EU milk

production from dairy cows. (FAOSTAT, 2013)

Figure 1 - Development of organic milk production from dairy cows in EU-28, 2007-2014. Source: FiBL-AMI surveys 2009-2015

Organic farming distinguishes itself from conventional farming by adopting a system aimed at producing food with minimal harm to ecosystems, animals or humans. (Seufert, Ramankutty, & Foley, 2012) Organic milk production reduces pesticide use and mineral surplus in agriculture. (Cederberg & Mattsson, 2000) The EU stimulates organic farming by supporting the organic producers with payments aimed directly at the organic economic system. Besides stimulation from the EU, the growth of organic dairy farming is caused by increased demand of consumers for bio-foods. (Malá, 2011)

2.4 2.4 2.6 2.8 3 3.1 3.3 3.8 0 1 2 3 4 5 2007 2008 2009 2010 2011 2012 2013 2014 Millio n to n n es

5 Nevertheless, there is much critique on organic farming, because it is less efficient than conventional farming. The yields of organic farming have been found to be low compared to conventional systems. This lower yield of organic farming is identified as the main drawback of organic farming (Nemecek et. al. 2011). For dairy farming the yield is expressed in

kilograms of milk produced per cow. Specifically, for the organic dairy industry, the yield is around 10% lower than the yield of the conventional dairy industry. (Nieberg & Offermann, 2000)

For 30 years, the EU has imposed production quotas on the dairy industry. These quotas were introduced in 1984, after years of significant overproduction of milk and milk products. (Marquer, 2015) Each Member State has two quotas, one for deliveries to dairies and one for direct sales at farm level. These national quotas are distributed among the individual

producers in the country. If a farm produces more milk than allowed for by its quota, a surplus levy (or “superlevy”) is imposed upon the farm. (Marquer, 2015)

Since 2004 there has been a loosening of these milk quotas. Several articles explore the effect of the phasing out or abolishment of the milk quota on conventional milk production. (Colman, 2002; Lips & Rieder, 2005) However, not much research has been done on the effect of phasing out the EU milk quota on organic dairy farms. It is interesting to see what the effects of loosening the milk quota are on the productivity of organic relative to

conventional milk production.

In this study, developments in productivity of both conventional and organic dairy farms during the period 2004-2011 are analyzed and compared. Specifically, I aim at answering the question: how does the phasing out of the EU dairy quota affect productivity of organic relative to conventional dairy farming? Chapter 2 reviews relevant literature. Chapter 3 discusses the methodology. Chapter 4 gives data descriptions. Chapter 5 presents the main results. The limitations of this research and suggestions for further research are given in Chapter 6. Finally, Chapter 7 summarizes the main findings of this research.

6

2. Literature review

While there is a lack of reliable studies on the future of organic dairy farming after phasing out the EU milk quota (Breustedt et al., 2011), relevant articles have been found to give shape to this research. In this chapter, I briefly discuss three most relevant strands of literature. The first strand of literature provides methods for measuring productivity in dairy farming, the second strand presents comparisons of the conventional and organic dairy industries and the third strand explores literature examining the effects of phasing out the milk quota.

2.1 Measuring productivity in dairy farming

Productivity is an important indicator of competitiveness in the dairy farming industry. (Woodford, Greer, & Phillips, 2003) Productivity is the efficiency with which the inputs are used to produce outputs. (Hannula, 2002) Two common indicates of productivity are total productivity and total factor productivity (TFP). Total productivity is the ratio of total output to total inputs used in the production process, i.e. output per unit of input. It is commonly used as the measure of competitiveness at the business unit or at the national level. (Hannula, 2002)

Total productivity = (Craig & Harris, 1973)

Craig & Harris (1973) define total input as the sum of labor, capital, raw material and purchased parts and other miscellaneous goods and services.

TFP is the growth in output not explained by input growth and can be used to measure technical progress and technical efficiency. (Chen, Yu, Chang, & Hsu, 2008) In dairy farming, growth in TFP means that the amount of kilograms produced per cow, or the yield per cow, increases. Kalirajan, Obwona & Zhao (1996) look at TFP to measure technical progress and technical efficiency in Chinese agriculture before and after reforms. A Cobb-Douglas technology is assumed, with the output value of the firm as the dependent variable and area, land, labor, machinery, chemical fertilizer and a time trend as independent variables. The time trend captures the change in TFP. Sipiläinen & Oude Lansink (2005) use stochastic frontier analysis to measure technical change and technical efficiency in organic dairy

farming in Finland from 1991 until 1999. They also use a production function. The dependent variable is milk in kilograms and the independent variables are labor, land, energy, material,

Total output Total input

7 capital, livestock units, a time trend and an extra error term for technical efficiency. The error term for technical efficiency is predicted by the age of the farmer and the years of experience in organic dairy farming.

2.2 Comparisons of organic and conventional dairy farming

With the same inputs, the average milk output per cow of organic dairy farms is typically around 10% lower than the milk output per cow of conventional farms. (Nieberg &

Offermann, 2000; Sipiläinen & Oude Lansink, 2005) The cause of the yield gap between organic and conventional farming can be decomposed into two parts: technical progress and technical efficiency. (Kalirajan et al., 1996) Technical efficiency means that the inputs are used in the most efficient way. This implies the best practice techniques available are

implemented and the highest possible output, given the inputs, is reached. Technical progress means the development of the best practice techniques. If there are new techniques available, the highest possible output increases for unchanged inputs. In other words, technical

efficiency means that the production moves closer to the production possibility frontier and technical progress means the production possibility frontier itself shifts outwards. (Kalirajan et al., 1996) The production possibility frontier of organic methods is lower than the

production possibility frontier for conventional methods. Organic dairy farms could have produced 5.3% more with conventional technology. (Kumbhakar, Tsionas, & Sipiläinen, 2009)

Little is known about the production practices of the organic dairy sector, not in small part because the industry is so new. (Mayen, Balagtas, & Alexander, 2010) Nonetheless, several articles explain the causes for the lower yield of organic (dairy) farming. For organic farming, there are strict rules for the use of chemical fertilizers, chemical pesticides and antibiotics. Therefore, yield limiting factors, such as nutrient limitations, and pests and diseases generally play a larger role in organic agriculture. (de Ponti, Rijk, & Van Ittersum, 2012) Strict rules for organic farming cause organic dairying to be accompanied by higher production costs. (McBride & Greene, 2009) For example, livestock should have access to open air or grazing area and feed must be organic. (The Council of the European Union, 2007)

Experience and knowledge gained over time, good management practices and learning by doing are important determinants of yields in organic dairy farming. (de Ponti et al., 2012; Sipiläinen & Oude Lansink, 2005) Additionally, organic yields depend more on knowledge and good management practices than conventional yields. (Seufert et al., 2012) Technical

8 efficiency is significantly positively related to experience in organic farming. (Sipiläinen & Oude Lansink, 2005) The average experience in years of organic farmers is small relative to conventional farmers. Therefore, the lower average experience of organic farmers could also be a cause for the lower yield of organic farming.

2.3 Phasing out the EU milk quota

The lower yield of organic milk production has been established in the literature quoted above. Hence, it is interesting to see if organic farms are moving towards or away from conventional farms in terms of productivity, after the loosening of the milk quota. A general effect of the loosening of the EU milk quota has been that milk production has increased. Figure 2 shows the loosening of the milk quota and the increase in milk production since 2004.

Figure 2 - Evolution of EU milk deliveries and direct sales versus quota. (European Commission, 2015)

A distinction has been made between direct sales and deliveries. Total EU Direct Sales comprises the raw milk that is sold directly from the farm and Total EU Deliveries comprises all raw milk that is delivered to dairies for processing, storage and distribution of milk and milk products. All milk quota have a fat reference, the percentage of fat the quota are based on. If the producer delivers milk which contains more or less fat, he can deliver less or more milk respectively. This is the fat correction. (European Commission, 2004)

9 The effect of abolishing the EU milk quota on conventional dairy farming has been

analyzed in several articles. Lips & Rieder (2005) predict that output will increase by 3% after abolition of milk quota in 2015. Furthermore, they predict a minor increase in milk production in Denmark, Italy, Luxembourg, the Netherlands and Spain and a decline for Germany, Greece, Portugal and Sweden. Harvey and Colman (2002) also state that abolition of dairy quota will have relatively minor effects on the EU and the rest of the world.

Breustedt et al. (2011) examined the effect of the abolition of the milk quota on organic dairy farms. It is stated that abolishment of the EU milk quota might considerably decline competitive advantage of organic dairy farms and as a consequence, they might lose market share. They determined whether it was optimal to be organic or conventional for 1300 Bavarian dairy farms under different policy scenario’s. In a situation with a milk quota, organic farming is optimal for 25.7% of the dairy farms in their sample, opposed to 3.4% in a scenario without milk quotas. This suggests that milk quota play a crucial role in maintaining competitiveness of organic dairy farming and that an abolition of milk quota is expected to be accompanied by a fall in organic dairy farms. However, the analysis is based on input–output observations from only one year.

The articles discussed above all focus on the abolition of the milk quota in 2015, but total EU milk production has not reached its quota for years. (European Commission, 2015) Because some countries have still produced up to their quota until 2015, it is interesting to study the effects of the abolishment of the milk quota. On the other hand, Figure 2 implies that the loosening of the milk quota in 2004 has already had a great effect on the milk

production in the EU. This stresses the importance of determining the effects of the loosening of the milk quota during the period 2004-2011. Based upon the article of Breustedt et al. (2011), which states that competitiveness of organic dairy farming will be reduced by

abolishing EU milk quota, the hypothesis is that productivity of relative to conventional dairy farming decreases after relaxing the milk quota in 2004.

10

3. Methods

In this chapter, I will explain how to measure productivity for both organic and

conventional dairy industries in the EU during the period 2004 until 2011. Total productivity and TFP will be measured. To control for differences between countries, only EU countries for which data is available on both organic and conventional farms will be compared. Otherwise, different costs of living and purchasing power could make the comparison less meaningful. (Nieberg & Offermann, 2000)

3.1 Measuring total productivity

Total productivity = (Craig & Harris, 1973)

Total output is expressed in euros and defined as the farm’s revenue. Total input is also expressed in euros and defined as the sum of total specific costs (feed, fertilizers and other livestock specific costs), overheads (supply costs linked to production activity but not linked to specific lines of production), depreciation of capital assets over the accounting years and external factors (wages, rent and interest paid). (FADN, 2010)

3.2 Measuring Total Factor Productivity

I assume that farms adopt a Cobb-Douglas production function with a time trend to measure TFP, as was also done by Kalirajan et al. (1996) and Sipiläinen et al. (2005). The change in TFP is the change in the milk yield per cow that cannot be attributed to a change in the inputs. Two Cobb-Douglas production functions are needed: one for conventional dairy farming and one for organic dairy farming. This allows for the possibility that organic and conventional farms make use of different production technologies.

The dependent variable (output) is measured as the average farm yield in kg/cow. The independent variables (inputs) include labor, area, the value of the machinery, the number of cows and the costs of feed and livestock specific costs. A time trend is added to capture the change in yield that can be attributed to a change in technical efficiency and technical progress. To estimate the function using ordinary least squares (OLS) linear regression, the Cobb-Douglas function has to be transformed to a linear form. This is done by taking the natural logarithms of the dependent variable and the inputs. The Cobb-Douglas production function then takes the following form:

Total output Total input

11 ln 𝑄𝑖𝑡 = 𝛽0 + 𝛽1ln 𝐿𝑖𝑡+ 𝛽2ln 𝐻𝐴𝑖𝑡+ 𝛽3ln 𝑀𝐴𝐶𝐻𝑖𝑡+ 𝛽4ln 𝐷𝐶𝑖𝑡+ 𝛽5ln 𝐹𝑆𝐶𝑖𝑡+ 𝛽6𝑡

+ 𝜀𝑖𝑡 (1)

i = country t = year

𝑄 = milk yield in kilograms/cow 𝐿 = labor input in full-time people 𝐻𝐴 = area in hectares

𝑀𝐴𝐶𝐻 = value of machinery in euro’s 𝐷𝐶 = the number of dairy cows

𝑇𝑆𝐶 = total specific costs (costs for feed and livestock specific costs) in euro’s

Regression (1) is estimated for both conventional and organic methods. The coefficient for the time trend, β6, demonstrates change in TFP. To determine the significance of changes

in TFP, the hypotheses defined below in equation (2) are tested at the 5% significance level. 𝐻0: 𝛽6 = 0 𝐻𝐴: 𝛽6 ≠ 0 (2)

To compare the change in TFP for organic and conventional industries, the data for organic and conventional methods is joined and sorted by type; conventional or organic. A dummy variable, ‘organic’, which is 1 for organic methods and 0 for conventional methods, and an interaction term between time and organic, ‘org*t’, are added to equation (1). The new regression is depicted in equation (3).

ln 𝑄_𝑖𝑡 = 𝛽₀ + 𝛽₁ln 𝐿_𝑖𝑡 + 𝛽₂ln 𝐻𝐴_𝑖𝑡+ 𝛽₃ln 𝑀𝐴𝐶𝐻_𝑖𝑡+ 𝛽₄ln 𝐷𝐶_𝑖𝑡+ 𝛽₅ln 𝐹𝑆𝐶_𝑖𝑡+ 𝛽₆𝑡 + 𝛽7𝑜𝑟𝑔𝑎𝑛𝑖𝑐 + 𝛽8𝑜𝑟𝑔 ∗ 𝑡 + 𝜀𝑖𝑡 (3)

If the interaction term is found to be significant at the 5% significance level, it can be stated that change in TFP is different for organic and conventional farms during 2004-2011. (Institute for Digital Research and Education, 2012) The hypotheses are shown in equation (4).

12

4. Data

4.1 Dataset Farm Accountancy Data Network

The dataset was obtained from the EU Farm Accountancy Data Network (FADN). The FADN is an instrument for evaluating the income of agricultural holdings and the impacts of the Common Agricultural Policy. Member states collect yearly accountancy data from a sample of agricultural holdings by a national survey. The bookkeeping principles are the same in all countries, which makes the FADN a harmonized source of microeconomic data. The FADN aims to provide representative data among three dimensions: region, economic size and type of farming. (FADN, 2010) Data is only available on country level; farm level data is confidential. For this research, a comparison is made between organic and

conventional farming. A dataset including data on organic farming was available for the period 2004 until 2011. The organic farms in the dataset satisfy EU regulation for organic farms and have completed the two year conversion period to becoming an organic farm.

Data on dairy farming is displayed as an average of all farms in the survey per year, per country and for organic and conventional farms. Data includes farm output, labor input, farm size, number of cows and other livestock, the milk yield, the value of machinery, the amount of taxes and subsidies, money spent on fertilizers etc. Also an estimate of the number of farms participating in the survey is displayed per category. For Germany, Denmark, France, the Netherlands, Austria, Poland, Finland, Sweden and the United Kingdom data was available on both conventional and organic farming. Only these countries have been used in the analysis, to prevent differences between countries to bias the comparison between organic and conventional methods. In 2013, these countries produced 71% of total EU cow’s milk production. (Eurostat, 2015)

In these countries altogether, between 45.000 and 85.000 conventional dairy farms and between 3485-8600 organic dairy farms have participated in the survey (only an

approximation of the number of participating farms is shown in the dataset). Because only the farm averages per year and per country are shown and only the countries for which data is available on both organic and conventional methods have been used, there are 71

13 4.2 Milk prices

The availability of accurate statistics on the organic market across Europe remains limited. (Schaack & Willer, 2010) Organic milk prices for the EU or the individual countries were not readily available in one database. Therefore, organic milk prices have been retrieved for each country individually. Only for Sweden, Finland and Poland it was not possible to retrieve the organic milk prices for the period 2004-2011.

4.3 Descriptive statistics

In the analysis, two outputs are distinguished (milk yield and revenue) and five inputs (labor, area, machinery, dairy cows and total specific costs). The milk yield is measured in kilograms of milk produced per cow per year. Farmhouse consumption and farm use (distributed to animals) are included. Revenue, machinery and total specific costs have been expressed in euros. Revenue is the total farm output plus the change in valuation of livestock minus the purchases of livestock. Machinery includes the value of machines, cars and lorries, tractors and irrigation equipment (except when of little value or used only during one year). Total specific costs include costs on feed, expenditure on the use of common grazing land, veterinary fees, reproduction costs and milk tests. Labor, area and the dairy cows are

measured in physical inputs. Labor is the total labor input of the holding expressed in annual work units or full-time person equivalents. Area is measured in hectares and includes owned and rented occupied land. ‘Dairy cows’ covers the number of female bovine animals which have calved and are held principally for milk production for human consumption. (FADN, 2010) Table 1 presents the descriptive statistics.

Table 1 - Descriptive statistics - pooled, conventional and organic farm data for 2004-2011

All farms n=142 Conventional n=71 Organic n=71

Mean Std Dev Mean Std Dev Mean Std Dev

Output Milk yield (kg/cow) 6,743.0 1,358.0 7,256.5 1,132.8 6,227.5 1,376.4 Revenue (€) 186,916 156,595 189,885 144,965 183,949 168,410 Input Area (ha) 71.39 42.41 65.34 32.96 77.44 49.62 Labor (full-time 1.95 0.34 1.94 0.27 1.97 0.40

14 people) Machinery (€) 96,297 71,453 94,504 64,316 98,090 78,365 Dairy cows 53.72 38.26 55.89 37.67 51.54 38.98 Total specific costs (€) 76,201 80,421 80,822 76,283 71,580 84,647

It can be seen from Table 1 that the average milk yield is higher for conventional than for organic dairy farms. Also the average revenue is higher for conventional farms. Organic farmers use, on average, more land. The average labor used is approximately the same for conventional and organic farms. Conventional farmers spend less on machinery than organic farmers, but use more dairy cows. The total specific costs are on average higher for

conventional than for organic farms, but the standard deviations for the total specific costs are high for conventional as well as for organic farms.

15

5. Results

This section exhibits results and analyses of the data for the period after the relaxing of the milk quota. It is divided into two parts. The first part presents a data analysis to illustrate developments in total productivity. The second part presents the results of the regressions to analyze TFP. This part also includes some additional graphs obtained from the data, to clarify the regression results.

5.1 Total productivity

Figure 3 reports the total productivity of organic and conventional farms, defined as total output (revenue) divided by the total inputs. Total productivity is generally higher for

conventional than for organic farms. In 2011, total productivity is slightly lower than in 2004 for both conventional and organic industries. For organic industries, this difference is larger. However, Figure 3 depicts that total productivity and trends in total productivity since the loosening of the milk quota are very much alike.

Figure 3 - Total productivity (Total output/Total input) , data source: FADN

Because total output is defined as the revenue, total productivity is highly influenced by price fluctuations, portrayed in Figure 4. Price fluctuations are alike for both organic and conventional milk. During 2006-2008, there is an increase in the milk price, followed by a sharp decrease in the milk price in 2008-2009, due to the dairy milk crisis. (Petrick & Kloss, 2013) The organic price is on average 22.54% higher than the conventional price. If we would control for this price difference, organic total productivity would be significantly lower than conventional total productivity.

0.9 0.95 1 1.05 1.1 1.15 2004 2005 2006 2007 2008 2009 2010 2011 T o tal o u tp u t/to tal in p u t Conventional Organic

16

Figure 4 - Average milk price (for Denmark, the Netherlands, Germany, Austria, France and the United Kingdom)

5.2 Total factor productivity

Table 2 shows the estimation results of regression (1) (p. 11) for both the conventional and the organic industry.

Table 2 - Coefficient estimates and their significance

Conventional farming Organic farming

Coefficient Estimate t-value P-value Estimate t-value P-value βo (constant) 34.1579*** 4.63 0.000 41.9605*** 4.15 0.000 β1 (area) -0.0721** -2.09 0.040 0.0985** 2.10 0.040 β2 (labor) -0.3722*** -4.35 0.000 -0.3515*** -3.62 0.001 β3 (machinery) 0.0945** 2.46 0.016 0.1328*** 4.64 0.000 β4 (dairy cows) -0.3107*** -8.15 0.000 -0.3449*** -9.76 0.000 β5 (total specific costs) 0.3795*** 7.66 0.000 0.2769*** 9.84 0.000 β6 (time trend) -0.0143*** -3.81 0.000 -0.0182*** -3.57 0.001 F (6,64) 58.77 118.993 R2 0.8577 0.9157 Significance level: * at 10%; ** at 5%; *** at 1%

All coefficients in both models are significant at the 5% significance level. In the models, the dependent and independent variables, except for the time trend, are expressed in

logarithms. Therefore the interpretation of the results is as follows: if for example area

increases by one percent, milk yield is expected to increase by β1 percent. The time trend β6is

interpreted differently: for each year added, the milk yield increases by around1 β6 percent.

The quantities of labor and dairy cows, have a negative effect on the milk yield for both conventional and organic farms. Machinery and total specific costs have a positive influence.

1 _{For each year added, the milk yield increases by a factor of 𝑒}𝛽₆

, which is close to 1 + 𝛽₆ for low values of 𝛽₆

25 30 35 40 45 2004 2005 2006 2007 2008 2009 2010 2011 Milk p ric (€ /1 0 0 k g ) Conventional Organic

17 Area has a negative effect on the milk yield for conventional farms and a positive effect for organic farms.

The time trend is negative for both models, which means TFP decreases over the time period 2004-2011 and thus both organic and conventional farms have become less productive in terms of TFP since the loosening of the milk quota. For conventional farms, with each year added, milk yield decreased with 0.0143%. For organic farms, milk yield decreased with 0.0182% with each year added. Accordingly, changes in TFP since the relaxing of the quota are minor. Over the period 2004-2011, conventional TFP decreased with 0.1144% and organic TFP decreased with 0.1456%.

To examine if the difference between the coefficients for the time trend of the conventional and organic model is significant, regression (2) (p. 11) has been run. The estimation results are shown in Appendix 1. These results prove that there is no significant difference between the coefficient β6 for organic or conventional industries. Therefore, the

null-hypothesis, which is that change in TFP is the same for organic and conventional methods, cannot be rejected.

Figure 5 - Average milk yield (kg/cow), data source: FADN

As demonstrated in Figure 5, the average milk yield has increased for both industries in the period 2004-2011, which seems contradictory to the fact that TFP has decreased. This can be explained by the fact that TFP is defined as the output over the regression coefficient-weighted sum of inputs. TFP can thus be constant or negative if an increase in yield is accompanied by an increase in inputs. The figures below illustrate the relative use of inputs for 2004-2011 for conventional (Figure 6) and organic (Figure 7) farms. The quantity in 2004 equals 1. Average inputs have increased for both organic and conventional farms.

5500 6000 6500 7000 7500 8000 2004 2005 2006 2007 2008 2009 2010 2011 Milk y ield ( k g /co w ) Conventional Organic

18

Figure 6 - Input growth in conventional farming, data source: FADN

Figure 7 - Input growth in organic farming, data source: FADN

0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2004 2005 2006 2007 2008 2009 2010 2011 Gr o w th ( ref er en ce 2 0 0 4 ) Labour Hectares Dairy cows Value machinery 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2004 2005 2006 2007 2008 2009 2010 2011 Gr o w th ( ref er en ce 2 0 0 4 ) Labour Hectares Dairy cows Value machinery

19

6. Discussion

This study shows that trends in productivity are alike for organic and conventional dairy farming during the years 2004-2011; the period after the loosening of the milk quota. I have determined trends in total productivity and total factor productivity. The difference in total productivity between 2004 and 2011 is minor for both conventional and organic systems. Fluctuations in total productivity are primarily caused by strong fluctuations in milk prices. Especially the effect of the 2009 dairy crisis is clearly reflected in total productivity

fluctuations. Figure 3 demonstrates that total productivity for organic and conventional methods are alike and that fluctuations are parallel.

Even though milk yield increased during 2004-2011, changes in TFP are negative. The relaxing of the milk quota allowed farmers to augment their milk production. This probably incentivized farmers to invest in their means of production, causing strong input growth in 2004-2009. This explains negative changes in TFP despite the increase of the milk yield. TFP changes are significant but small for both organic and conventional systems. The difference between changes in TFP between organic and conventional farms is not significant.

The outcome is relevant for policy makers. The EU stimulates organic dairy farming. If phasing out the dairy quota reduces competitiveness of organic dairy farms, extra measures might be desired to increase competitiveness of organic dairy farming. However, I showed that organic relative to conventional competitiveness in dairy farming is not reduced.

Breustedt et al. state that policy makers will have to reconsider organic support policies after abolition of the milk quota, if promotion of organic agriculture remains a political priority. They illustrated that abolishing the EU milk quota results in loss of competitive advantage of organic relative to conventional dairy farms. Our results show no loss of competitiveness for the organic dairy industry since the relaxing of the milk quota, indicating that abolishing the milk quota will at the most have a moderate effect on competitiveness of the organic dairy industry, according to the results of this study.

Only data on country-level was available. Most studies examining the productivity and profitability of (dairy) farms use data on farm-level. Using data on farm-level allows for stochastic frontier analysis (SFA). SFA computes overall and input specific technical

efficiencies. (Sipiläinen & Oude Lansink, 2005) It can be used to distinguish changes in TFP into changes in technical efficiency and technical change, which I have not been able to do. For future research, it is recommended to use data on farm-level and to use SFA to measure changes in TFP.

20 If data on farm-level can be used, it would also be interesting to look at the quota for each farm specifically and incorporate the quota into the Cobb-Douglas production function, to control for the effect of the quota on the milk yield. There was data on the quotas for each individual country, but the data did not distinguish between organic and conventional quota. Also, the data used is in this research is based on the average organic and conventional farm of each country and data on quota was only available for the country as a whole. Therefore, the quota have not been incorporated in the regression.

21

7. Conclusion

Loosening of the EU milk quota started in 2004, causing EU milk production in 2007 to be under its quota for the first time in 23 years. I analyzed the effect of phasing out the milk quota on productivity of organic relative to conventional dairy farms in the EU during the period 2004-2011. Trends in total productivity and TFP were determined and compared for the organic and conventional dairy industry. Total productivity decreased slightly during this period for both industries. Fluctuations in total productivity are primarily caused by price fluctuations. Trends in total productivity are similar for organic and conventional methods. Milk yield increased during 2004-2011. However, changes in TFP are negative for both organic and conventional farms. This can be explained by input growth in 2004-2009. The difference between changes in TFP between organic and conventional farms is not significant.

The results show the same trends in total productivity and TFP for organic and

conventional dairy farms in 2004-2011 and thus their relative productivity remains the same. This contradicts our hypothesis, stating that loosening the milk quota will have a relatively negative effect on productivity of organic as compared to conventional dairy farming.

22

References

Breustedt, G., Latacz-Lohmann, U., & Tiedemann, T. (2011). Organic or conventional? optimal dairy farming technology under the EU milk quota system and organic subsidies.

Food Policy, 36(2), 223-229.

Cederberg, C., & Mattsson, B. (2000). Life cycle assessment of milk production — a comparison of conventional and organic farming. Journal of Cleaner Production, 8(1), 49-60.

Chen, P., Yu, M., Chang, C., & Hsu, S. (2008). Total factor productivity growth in china's agricultural sector. China Economic Review, 19(4), 580-593.

Colman, D. (2002). Phasing out milk quotas in the EU.

Craig, C. E., & Harris, R. C. (1973). Total productivity measurement at the firm level. Sloan

Management Review, 14(3), 13.

de Ponti, T., Rijk, B., & Van Ittersum, M. K. (2012). The crop yield gap between organic and conventional agriculture. Agricultural Systems, 108, 1-9.

Commission regulation (EC) no 595/2004 of 30 march 2004 laying down detailed rules for applying council regulation (EC) no 1788/2003 establishing a levy in the milk and milk products sector, (2004).

European Commission. (2015). Milk quota expiry and market prospects for the dairy sector [powerpoint slides].

Eurostat. (2015). Milk and milk product statistics. Retrieved from

http://ec.europa.eu/eurostat/statistics-explained/index.php/Milk_and_milk_product_statistics

FADN. (2010). Farm accounting data network an A to Z of methodology. Unpublished manuscript.

FAOSTAT. (2013). Milk, production share by region. Retrieved from

http://faostat3.fao.org/browse/Q/QL/E

Hannula, M. (2002). Total productivity measurement based on partial productivity ratios.

International Journal of Production Economics, 78(1), 57-67.

Institute for Digital Research and Education. (2012). Stata FAQ - how can I compare regression coefficients between 2 groups? Retrieved from

http://www.ats.ucla.edu/stat/stata/faq/compreg2.htm

Kalirajan, K. P., Obwona, M. B., & Zhao, S. (1996). A decomposition of total factor productivity growth: The case of chinese agricultural growth before and after reforms.

23 Kumbhakar, S. C., Tsionas, E. G., & Sipiläinen, T. (2009). Joint estimation of technology

choice and technical efficiency: An application to organic and conventional dairy farming. Journal of Productivity Analysis, 31(3), 151-161.

Lips, M., & Rieder, P. (2005). Abolition of raw milk quota in the european union: A CGE analysis at the member country level. Journal of Agricultural Economics, 56(1), 1-17. Malá, Z. (2011). Efficiency analysis of czech organic agriculture. E M Ekonomika A

Management, 1, 14-28.

Marquer, P. (2015). Milk and milk products - 30 years of quotas. Retrieved from

http://ec.europa.eu/eurostat/statistics-explained/index.php/Milk_and_milk_products_-_30_years_of_quotas

Mayen, C. D., Balagtas, J. V., & Alexander, C. E. (2010). Technology adoption and technical efficiency: Organic and conventional dairy farms in the united states. American Journal

of Agricultural Economics, 92(1), 181-195.

McBride, W. D., & Greene, C. (2009). Costs of organic milk production on US dairy farms.

Applied Economic Perspectives and Policy, 31(4), 793-813.

Nieberg, H., & Offermann, F. (2000). Economic performance of organic farms in europe. Petrick, M., & Kloss, M. (2013). Exposure of EU farmers to the financial crisis. Choices–The

Magazine of Food, Farm, and Resource Issues, 28(2)

Pierce J., T. O. (2015). Introduction to organic dairy farming. Retrieved from

http://articles.extension.org/pages/18325/introduction-to-organic-dairy-farming

Schaack, D., & Willer, H. (2010). Development of the organic market in europe. Training

Manuals for Organic Agriculture, , 141.

Seufert, V., Ramankutty, N., & Foley, J. A. (2012). Comparing the yields of organic and conventional agriculture. Nature, 485(7397), 229-232.

Sipiläinen, T., & Oude Lansink, A. (2005). Learning in switching to organic farming. NJF

Report, 1(1), 169-172.

Council regulation (EC) no 834/2007 of 28 june 2007 on organic production and labelling of organic products and repealing regulation (EEC) no 2092/91, (2007).

Woodford, K. B., Greer, J., & Phillips, A. (2003). Searching for productivity and competitive advantage on new zealand dairy farms. Proceedings of the 14th International Farm

Management Congress. England: International Farm Management Association,

24

Appendix 1

Table 3 shows the estimation results of regression (1) for both the conventional and the organic industry.

Table 3 - Coefficient estimates and their significance

Coefficient Estimate t-value P-value βo (constant) 33.3350*** 4.78 0.000 β1 (area) 0.0087 0.31 0.754 β2 (labor) -0.2967*** -4.89 0.000 β3 (machinery) 0.1466** 7.26 0.016 β4 (dairy cows) -0.3063*** -11.97 0.000 β5 (total specific costs) 0.2923*** 11.34 0.000 β6 (time trend) -0.0139*** -3.98 0.000 β7 (organic) 13.5044 1.28 0.203 β8 (organic*time) -0.0068 -1.29 0.200 F (8,133) 163.75 R2 0.9042 Significance level: * at 10%; ** at 5%; *** at 1%

The interaction term β8 (organic*time) is not significant at the 5% significance level. Thus,