The Future of European Union Dairy Policy

Catherine Benjamin, Alexandre Gohin and Herv Guyomard

INRAESR, Rennes, France.
Even if the quota regime has been extended until 200708 as part of the Agenda 2000 reform adopted in March 1999, the European Union (EU) dairy sector is currently experiencing large uncertainty over future policy. This paper examines EU dairy policy issues and assesses their likely implications. It mainly addresses the central question of dismantling the quota and intervention support mechanisms. Simulation results illustrate how three factors (marginal costs of production, import tariffs on dairy products and compensatory payments granted to dairy producers) may affect the outcomes of a quota elimination scenario. Mme si le rgime des quotas a t prolong jusquen 20072008 loccasion de la rforme Agenda 2000 de mars 1999, le dbat sur la ncessit dune rforme plus importante de la politique laitire europenne nest pas clos. Dans ce papier, nous analysons les diffrentes contraintes auxquelles lUnion europenne doit faire face et ses implications pour le secteur laitier communautaire. Lattention est centre sur les consquences dune suppression du rgime des quotas et du mcanisme de lintervention. Les rsultats de simulation montrent comment les consquences dun tel scnario dpendent de trois facteurs, i.e., les cots marginaux de production du lait, la protection tarifaire sur les importations europennes de produits laitiers et les mesures de compensation accordes aux producteurs de lait.

INTRODUCTION Supply management policies are often advocated as a possible response to a situation of overproduction that protects producers incomes without boosting budgetary costs. This was effectively the rationale for the introduction of milk quotas in the European Union (EU) in 1984. Even if the quota regime has been extended until 200708 as part of the Agenda 2000 reform of the Common Agricultural Policy adopted in March 1999, the EU dairy sector is still nowadays experiencing large uncertainty over future policy. This paper surveys the main economic characteristics of the EU dairy sector, along with policy arrangements such as the Uruguay Round Agreement on Agriculture commitments and Agenda 2000 measures.

After analyzing current dairy policy issues, some thoughts on the central question of an eventual dismantling of quotas are offered. THE EU DAIRY SECTOR: TIME FOR RADICAL POLICY CHANGE? The first principle underlying the current EU dairy policy is the management of the markets for dairy products in order to secure product prices that permit domestic milk producers to obtain the target price for milk. This involves the annual fixing of both a target price for farm milk and intervention prices for butter and skimmed milk powder (SMP). The public purchase of butter and SMP is triggered when the market price falls below 92% of the intervention price. This also involves the payment of aids for the domestic disposal of some dairy products

Canadian Journal of Agricultural Economics 47 (1999) 91101

91

92

CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS

(SMP for animal feed, butter for pastry and ice cream manufacturers, butter for non-profit making organizations and milk products supplied to school children) and for the private storage of butter and cream, SMP and certain cheeses. This finally includes the payment of variable export refunds for a large set of dairy products (butter, SMP, whole milk powder (WMP) and cheese) and the imposition of tariffs on imports of dairy products (CAP Monitor 1999). Faced with the problem of achieving a better balance between supply and demand and the induced problem of an increased budgetary cost, the EU introduced the socalled milk quota system in 1984. The quota system was extended until 2000 as part of the 1992 Common Agricultural Policy reform package. It is proposed to continue until 200708 as part of the Agenda 2000 Common Agricultural Policy reform of March 1999. Economic characteristics of the EU dairy sector Since 1984, dairy yields in the EU have continuously increased (+22% over the 12-year period 198597, from 4460 kilograms to 5446 kilograms per cow for the EU10). With increasing yields and the quota system, there has been a decline in the number of dairy cows from the mid-1980s onward (more than 20% between 1985 and 1997). There has been a drastic decline in the number of dairy farms (59% over the 12-year period 198496, from 1,621,000 to 668,000 for the EU10). Although there is a common trend in all EU member states toward concentration and larger dairy farms, national holding structures are still very different. Whereas Austria and southern countries have an average of 1019 cows per holding, the predominant structure in eastern Germany, Italy and the United Kingdom corresponds to an average of 50100 cows (USDA 1998). Structure disparities both within and between member states imply that it is particularly difficult for the EU to modify the current dairy policy: firstly, production costs and gross margins are very heterogeneous and, secondly, smaller

and less efficient dairy farmers contribute to proper management of the landscape and nature in rural areas, particularly in less favored areas, which is also an objective of the agricultural and rural policy for several EU countries. More than 90% of EU milk output is delivered to dairies. The EU produces roughly 12 million tonnes more dairy products (expressed in whole milk equivalent) than the domestic market can absorb. This surplus has to be exported thanks to subsidies, which represent about 45% of the dairy policy budgetary cost (1.605 billion Ecus over a total of 3.635 billion Ecus in 1996). In addition, about 10% of domestic consumption can be sustained only by butter and SMP internal subsidies, which represent more than 40% of the dairy policy budgetary cost (1.508 billion Ecus in 1996). After a long period of decline since the mid 1970s, per capita consumption of butter in the EU15 has stabilized at around 4.7 kilograms per head with, however, very large disparities across member states (less than 0.5 kilograms per head in Spain and about 8.3 kilograms per head in France). Butter manufacture still absorbs one-third of the total milk produced in the EU, and 30% of domestic consumption of butter is subsidized. The share of SMP used in animal feed is continuously decreasing, although about 4045% of domestic consumption is subsidized. Domestic production and consumption of WMP are increasing. Less than 45% of production is used within the EU, mainly for human consumption in the food processing industry. EU production and consumption of cheese are on a long-term increasing trend since the mid-1970s. Per capita consumption levels exhibit similar disparities as butter does. Fresh dairy products are the third dairy products for which domestic consumption is going up. While output of drinking milk has been more or less stable over the past decade, other fresh dairy products (mainly cream, acidified milk and milk-based drinks) are on an upward trend. The EU is still nowadays the major world exporter of dairy products with more than 4045% of world trade (Table 1). Its

FUTURE OF EUROPEAN UNION DAIRY POLICY

93

Table 1. EU exports of dairy products (kt)a 1990 (12) 1991 (12) 1992 (12) 1993 (12) 1994 (12) 1995 (15) 1996 (15) 1997 (15) Butterb EUs share NZs share SMP EUs share NZs share WPM EUs share NZs share Cheese (unsubs.) EUs share NZs share
aIncluding bButter

260 32.1 330 35.4 na 502 56.7 na 450 na 50.7 na

322 36.9 253 28.8 na 618 57.0 na 483 56 51.8 na

242 32.8 33.6 390 39.1 13.9 581 55.4 28.4 465 62 50.5 13.6

201 24.5 32.7 283 30.6 11.1 588 53.8 27.9 524 80 52.9 12.8

154 20.9 39.1 138 16.4 15.3 586 51.2 30.3 510 84 47.9 12.9

228 27.0 38.9 371 30.8 12.4 596 46.4 28.4 521 63 48.7 15.9

192 24.4 40.7 224 23.3 15.7 538 46.4 25.8 500 na 45.9 17.5

222 25.5 44.5 279 30.9 18.2 569 44.4 27.3 453 na 41.0 23.4

food aid. and butteroil. na = Not available. Source: The Agricultural Situation in the European Union, 1998 Report (statistical annex), except the New Zealands share (Lattimore and Amor, 1998) and nonsubsidized exports (European Commission, April 1997, page 28).

market share is steadily decreasing since the mid-1980s, mainly in favor of New Zealand and Australia. EU market prices under pressure Two categories of EU dairy products, i.e., cheese and other dairy products, are subject to tight Uruguay Round Agreement on Agriculture constraints on the export side, which require that by the year 200001, export subsidies should be reduced by 36% and the volume of subsidized exports by 21% relative to the base period. Further commitments on subsidized export cuts are more than likely in the context of the next round of World Trade Organization (WTO) talks (Josling and Tangermann 1999). Since only specialty dairy exports can compete without export refunds in the current regime of high internal market prices, the EU is forced to make price cuts if its wants to maintain its share of world exports under the Uruguay Round Agreement on Agriculture and WTO conditions.

More severe WTO commitments on market access for imports, through reduced tariffs as well as augmented current and minimum access provisions, can also be expected. In addition, the EU faces the prospect of enlargement to central and eastern European countries (CEECs) where market prices of dairy products are still nowadays significantly lower compared with the EU15 (European Commission 1997; Fuller et al 1999). These are additional pressures for price cuts in the dairy sector. Assuming than the future of EU agriculture lies in world markets, the EU dairy policy should give dairy farmers and processors the greatest possible flexibility to react to world market signals. But a speedy removal of dairy restrictions and price support mechanisms raises the crucial question of the number of efficient EU dairy producers who could compete at world prices. This number closely depends on the compensation package since a quota elimination option is polit-

94

CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS

ically feasible only if dairy producers are at least partially compensated for the reduction in price support. A full compensation would induce a prohibitive budgetary cost. On the other hand, milk production would significantly decline (at the limit disappear) in the poorer dairy regions of the EU if compensatory payments are not modulated in their favor. The extent, nature and permanence of compensation and the subsequent distribution and justification of payments would play a large role in determining the EU dairy sector after quota elimination (OLeary and Fingleton 1998). AGENDA 2000 REFORMS AND CONSEQUENCES FOR THE EU DAIRY SECTOR In a general way, the Agenda 2000 Common Agricultural Policy reform will deepen (cereals and beef) and extend (dairy products) the 1992 MacSharry reform through further shifts from price support to direct payments. The dairy reform may be summarized as follows (European Commission 1999a, 1999b). Intervention prices for butter and SMP will be reduced by 15% in three equal steps starting on 1 July 2005. Support price cuts will be compensated by the introduction of yearly direct payments on a flat rate basis per tonne of quota and by an increase in milk quotas. The compensation scheme will include a national envelope, which member states are free to allocate according to objective criteria. Dairy producers will also receive some benefit from the beef compensation payments. Quotas will be increased in each member state by 1.5% in three steps starting on 1 April 2005. In addition, five countries (Greece, Italy, Spain, Ireland and Northern Ireland) will receive specific quota increases in two steps, i.e., 200001 and 200102. In the whole, this corresponds to an increase in milk quotas by 2.4%. The quota regime is thus extended, officially up to the marketing year 200506, in practice up to the year 200708. From the European Commission point of view, the dairy reform corresponds to a

cautious approach because expected market developments do not require extreme measures. However, the Berlin accord makes provision for a mid-term review of the quota regime in 2003, which could be interpreted as a signal for the phasing out and dismantling of quotas before the end of the next decade. The so-called London Club of Four (Denmark, Italy, Sweden and United Kingdom) has unsuccessfully campaigned for such a dismantling during the Agenda 2000 negotiations. It is more than likely that these four countries will continue to act in this direction. As discussed in the companion paper in this issue of the Journal by Fuller et al, the main effects in 2007 of the Agenda 2000 dairy measures would be: a decrease in EU farm milk prices by about 9.5%, butter and SMP prices falling by larger percentages than prices of other dairy products an increase in domestic consumption of all dairy products a decrease in EU exports of butter and SMP and an increase in EU exports of cheese and WMP a decline in SMP world prices and a rise in cheese world prices. Previous results rest on the assumptions that Uruguay Round commitments on market access and subsidized exports are unchanged at 2001 levels and that the EU can increase commercial, i.e., nonsubsidized, exports of cheese. Table 2 shows that the ultimate impact of the Agenda 2000 reform on farm milk price will crucially depend on the management of export subsidies and on substitution possibilities between intervention dairy products (butter and SMP) and other dairy products. The second column of this table shows that a 15% reduction in butter and SMP intervention prices leads to a 9.3% reduction in milk price at the farm gate in France, other dairy policy instruments being maintained unchanged at base period levels. The additional effect of a 15% decrease in unit export subsidies on cheese, WPM and fresh dairy products is to reduce the farm

FUTURE OF EUROPEAN UNION DAIRY POLICY

95

Table 2. Impacts of changes in dairy policy instruments on the French dairy sector Scenario 1 Assumptions LVDP intervention prices (%) HVDP unit export subsidies (%) Milk quota (%) Elasticity of transformation sa Results (% with respect to the base) Farm milk price HVDP price 15 0 0 0.5 9.26 1.75 Scenario 2 15 15 0 0.5 11.49 3.25 Scenario 3 15 15 +2 0.5 14.50 5.18 Scenario 4 15 0 0 0.9 10.10 2.67

LVDP = Lower-value dairy products, i.e., butter and SMP. HVDP = Higher-value dairy products, i.e., cheese, WMP and fresh dairy products. aThe elasticity of transformation s reflects substitution possibilities between LVDP and HVDP in the dairy industry. Source: Gohin and Guyomard (1999).

milk price by 2.2 points (11.5% instead of 9.3%) and the supplementary effect of a 2% quota increase is to reduce the farm milk price by 3 points (14.5% instead of 11.5%). The last column of this table shows that when substitution possibilities between intervention dairy products and other dairy products are increased, the effects of a reduction in butter and SMP prices on farm milk price are also increased. This last scenario illustrates the recent move from bulk commodities toward higher-value dairy products more closely focused on consumer needs. The previous analysis is confirmed by the FAPRIUMC study about the implications of the Agenda 2000 reform for EU agriculture (FAPRIUMC 1999). In an attempt to represent not only producer and consumer behaviors but also European Commission behavior, FAPRIUMC modelers clearly illustrate the trade-off faced by the European Commission, which may want to reduce the Euro value of export subsidies, in part because of smaller gaps between EU and world prices and simultaneously keep domestic market prices high enough . . . , so that market prices are likely to fall less than support prices. The maintenance of sufficiently high milk market prices is also politically essential to sustain the income of smaller, less efficient dairy farmers.

MILK QUOTA REMOVAL: A POLITICALLY AND ECONOMICALLY CONCEIVABLE SOLUTION? The Organisation for Economic Co-operation and Development (OECD) has analyzed the effects of removing milk quotas and dairy price support in Canada, Japan and the EU (OECD 1999; Thompson et al 1999). The EU milk market price would fall to the level of the average marginal cost of production which is estimated to be 25% lower that the milk price in the baseline scenario. High import tariffs would prevent domestic prices to fall to world levels. EU market prices would be determined through EU market clearing, and they would remain 23% higher than world prices at the end of the simulation period, i.e., 2004. This domestic price decline would not be sufficient to prevent a significant increase in milk production, which implicitly means that the current marginal cost of production is below the ex post equilibrium marginal cost. The FAPRIUMC study concurs that removal of dairy quotas and eliminating intervention support and domestic subsidy programs, export subsidy and tariff protection measures being maintained, would lead to a significant increase in milk production at substantially lower milk prices (FAPRIUMC 1998).

96

CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS

Both the OECD and FAPRIUMC studies paint a rather rosy picture of consequences of removal of dairy quota restrictions in the EU. Their conclusions rest on the crucial assumption that continuing tariff and non tariff protection will prevent domestic milk prices to fall to world levels and third country imports to significantly enter the EU market. Simulation results also depend on two key behavioral parameters, i.e., marginal cost and milk supply elasticity estimates. In addition, these studies do not include any income enhancement from a compensation program. Finally they ignore the related highly sensitive issue of the number of dairy farmers who could abandon milk production under a quota elimination scenario. MILK QUOTA REMOVAL: A SENSITIVITY ANALYSIS TO KEY PARAMETERS In order to illustrate the sensitivity of simulation results of a quota elimination scenario to the different parameters identified above, we use a computable general equilibrium (CGE) model of the French economy benchmarked to data for 1994 (Gohin 1998; Gohin and Guyomard 1999; Gohin et al 1999). The model is a static, single-country, multisector CGE model in the neoclassical tradition of Shoven of Whalley (1984). It highlights agricultural and food sectors, with special attention given to Common Agricultural Policy instrument modeling. The various dairy policy instruments are thus explicitly modeled. Production quotas are represented by means of an additional factor of production that ensures the zero profit condition holds for dairy farming. The mixed-complementarity approach is used to model the intervention price mechanism prevailing for butter and SMP so that the price regime is endogenously determined as part of the model solution. When the producer price is equal to the intervention price, the unit export subsidy is strictly positive, and the equilibrium variable ensures that the market clears. When the producer price is greater than the intervention price, the unit export subsidy equals zero.

For the various dairy products that are not regulated by the intervention system, prices adjust to ensure market equilibrium and the unit export subsidy is an exogenous political variable. Consumption subsidies are modeled by means of ad valorem subsidies, and fixed tariffs are applied on imports from nonEU member states. Two export and import zones are distinguished, i.e., the rest of the European Union (RoEU) and the rest of the world (RoW). This allows us to consider policy scenarios that apply at the EU level. Experiments are detailed in Table 3, and results are summarized in Table 4. Reference Experiments In experiment 1, dairy quotas, intervention support programs and domestic subsidy measures are eliminated. Import tariffs on all dairy products and unitary export subsidies on cheese, WMP and fresh dairy products (hereafter called higher-value dairy products) are maintained at base period levels. There are no compensation payments. Relative to the base year, this first experiment triggers a milk price decrease by 26.1% and a milk production increase by 4.3%. With the elimination of intervention support programs, butter and SMP domestic prices decline by a very large percentage (36.3%) while prices of other dairy products, which did not benefit from intervention outlets and continue to benefit from export subsidy measures, fall by 6.6%. Markets of higher-value dairy products attract additional milk supply. Without compensatory payments, dairy farmer incomes decrease by 21.3%, while dairy industry incomes increase by 5.5%. This last figure shows that the negative effects of dairy industry output price decreases are more than offset by the positive impacts of a lower milk price and an increased milk supply. Assumptions of experiment 2 are the same as those of experiment 1, except that export subsidy programs are eliminated not only for intervention dairy products but also for higher-value dairy products. Relative to experiment 1, this second scenario triggers more pronounced domestic price declines for

FUTURE OF EUROPEAN UNION DAIRY POLICY

97

Table 3. Experiment design No. 1. Experiment Quota removal Experiment design Removal of dairy quotas, elimination of intervention support and domestic subsidy programs, no compensatory payments, quota rent equal to 27.5% of the milk price Experiment 1 assumptions, and variable export subsidies on higher-value dairy products set to zero Experiment 2 assumptions, except that the dairy quota rent is equal to 17.5% of the milk price Experiment 2 assumptions, except that the dairy quota rent is equal to 37.5% of the milk price Experiment 2 assumptions, and import tariffs on all dairy products set to zero with an infinite import supply elasticity Experiment 2 assumptions, and import tariffs on all dairy products set to zero with a strictly positive import supply elasticity Experiment 2 assumptions, and production linked compensatory payments for a total amount equal to the income loss of dairy producers in experiment 2 (14,74 billion 1994 French francs) Experiment 2 assumptions, and area based compensatory payments for a total amount equal to the income loss of dairy producers in experiment 2 (14,74 billion 1994 French francs)

2. 3. 4. 5.

Quota removal and elimination of export subsidy measures Robustness to the quota rent value Robustness to the quota rent value No tariff protection with a perfectly inelastic import supply function No tariff protection with an elastic import supply function Production linked compensatory payments

6.

7.

8.

Area-based compensatory payments

higher-value dairy products (8.7% instead of 6.6%) and farm milk (28.2% instead of 26.1%) and less pronounced price declines for butter and SMP (33.3% instead of 36.3%). The decrease in milk output by 2.9%, together with the decline in milk price, reduce dairy farmer incomes by around 25% relative to the base. The dairy producer income decrease is limited relative to experiment 1, because the milk supply function is very elastic (e.g., Guyomard et al 1996). Dairy industry incomes now fall by more than 3% relative to the base, mainly because milk supply is reduced. To a large extent, this first set of experiments confirms OECD and FAPRIUMC conclusions about the effects on milk price of a quota elimination scenario. The size of the EU market is large enough so that domestic

milk prices do not fall by drastic amounts if dairy quotas and price support instruments are removed, at least under the assumption that continuing tariff protection prevents RoW imports to enter the EU dairy market so that domestic prices are determined through EU market clearing only. When export subsidy programs for higher-value dairy products are maintained, milk output increases relative to the base year (experiment 1). When these programs are simultaneously eliminated, the EU milk price decrease is sufficient to induce a milk production decline (experiment 2). In both experiments, the decrease in dairy producer incomes is too important to imagine a quota elimination policy without a compensation scheme to offset at least part of the decline in incomes.

98

Table 4. Experiment results (% with respect to the base)a Reference Exp. 1 21.25 26.13 +4.31 +5.49 36.25 10.70 73.79 +1.55 6.57 +8.11 16.54 +2.72 8.74 +0.13 21.92 +3.89 4.30 2.02 11.64 +1.80 13.51 +2.64 32.02 +6.33 15.86 31.89 (26577) +8.15 33.33 14.41 71.00 0.94 28.88 15.54 65.17 2.22 36.06 11.75 74.18 +0.18 34.61 39.95 (3785) 8.42 33.94 19.48 (779) 2.32 9.98 6.01 (5528) +4.60 3.06 5.41 0.04 37.24 9.93 +7.67 37.64 2.85 75.69 +1.96 22.17 +8.54 48.09 +11.42 24.90 28.21 2.88 17.91 18.86 4.83 32.05 37.51 0.35 38.51 38.40 33.59 27.70 30.07 8.80 20.36 53.24 +6.10 Exp. 2 Exp. 3 Exp. 4 Exp. 5 Exp. 6 Exp. 7 Exp. 8 +0.04 30.16 2.41 2.30 33.99 13.99 71.76 0.45 9.68 +0.61 24.06 +4.35 Dairy quota rent No tariff protection Compensation

Base year

59198 1 50188

21764

Dairy farmers Value added Price Production Dairy industry Value added LVDP Price Production M. from RoW Dom. demand HVDP Price Production M. from RoW Dom. demand

1 25788 40 21234

CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS

1 84907 803 75601

aQuantities in million 1994 French francs and prices set to one in the base year. In parentheses, import volumes in million 1994 French francs. LVDP = Lower-value dairy products (butter and SMP). HVDP = Higher-value dairy products. M. from RoW = French imports from the rest of the world, excluding other EU member states.

FUTURE OF EUROPEAN UNION DAIRY POLICY

99

Robustness to the Dairy Quota Rent Value Experiments 3 and 4 allow us to analyze the robustness of simulation results to the dairy quota rent value, initially estimated to 27.5% of the EU average milk price. The rent equals 17.5% of the milk price in experiment 3 and 37.5% of that in experiment 4. When the value of the rent decreases (respectively increases), the effects of a quota elimination policy on domestic prices are reduced (augmented) for farm milk, lower-value dairy products and higher-value dairy products. Although the milk price decline is lower in experiment 3 (18.9%) than in experiment 2 (28.2%), milk supply drops by a larger percentage (4.8% instead of 2.9%) due to the upward shift of the milk supply function induced by the lower value of the quota rent. The negative income effect of a lower milk production is more than offset by the positive impact of higher milk prices so that dairy farmer incomes increase in experiment 3 relative to experiment 2. Of course, they fall with respect to the base year, and a quota elimination option is still unlikely if it is not accompanied by a compensation package. Our second set of experiments shows not only that own price supply elasticities matter (i.e., the price responsiveness of milk supply in a no quota regime), but also that initial marginal costs of production and dairy quota rents influence the impact of the quota removal. No Tariff Protection Experiments In experiments 5 and 6, the tariffs on dairy products that protect the EU countries markets are set to zero. In both experiments, the standard Armington approach is modified to allow undifferentiated dairy imports to enter French and EU markets in a regime where tariffs are set to zero (Gohin et al. 1999). In experiment 5, we assume that EU import prices are equal to equilibrium world prices obtained in experiment 2 or, in other words, that the import supply function of dairy products from the RoW is perfectly elastic. This experiment brings about a dramatic deterioration in the expectations for the whole sector. In particular, French milk output decreas-

es by 33.6% with respect to the base. In experiment 6, we assume that France (and the EU) is a large country so that import prices of dairy products increase with imported volumes. The import supply function of dairy products from the RoW is thus less than perfectly elastic. Even if volume, price and income effects are less pronounced in this experiment relative to experiment 5, all dairy indicators still fall by catastrophic amounts. For example, French milk production decreases by 8.8% with respect to the base and French dairy producer incomes decline by 30.1%. In experiments 5 and 6, domestic price decreases are greater than in experiment 2 because the EU price system is no longer isolated from world prices so that French and EU prices are now determined through world market clearing. French and EU imports from the RoW increase by substantial amounts. In experiment 5, the value of French imports of higher-value dairy products represent more than 32% of total domestic demand while it corresponded to 1.1% only in the base year. Compensation Experiments A speedy elimination of dairy quotas and intervention support measures seems politically feasible only in the context of acceptable compensation, at least over a certain transition period. In experiment 7, compensatory payments offered to dairy producers are tied to production levels. In experiment 8, they are based on acreage declarations. To make the comparison meaningful, the total amount of compensatory payments is the same in both experiments, i.e., 14.74 billion 1994 French francs, which corresponds to the income loss of dairy producers in experiment 2. The coupled compensation package of experiment 7 does relatively little to support dairy producer incomes because it translates into a milk price decrease that nearly offsets the positive income effect of payments. Dairy producer incomes fall by 20.4% relative to the base year. In this experiment, production decisions of dairy farmers are based on an equivalent price that is the sum of the market price and the average compensatory

100

CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS

payment. As a result, despite the decrease by 53.2% in the milk market price, French milk output increases by 6.1% relative to the base year. This increase mainly benefits the dairy industry. For the latter, output price declines (37.6% for lower-value dairy products and 22.2% for higher-value dairy products) are more than offset by the positive income effects linked to both the milk price decrease and the milk output increase. As a result, dairy industry incomes raise by 7.7% with respect to the base year. By contrast, experiment 8 allows French dairy producers to maintain their incomes at base period levels. Relative to experiment 7, dairy producers now capture most of the benefits of the compensation package to the detriment of the dairy industry and final consumers. However, the cost of both compensation schemes is likely to be prohibitive, given the tight common agricultural budget. Any dairy quota elimination scenario should include compensation for a least part of the decline in dairy farmer incomes. The last set of experiments shows that any compensation package should be as decoupled as possible from current and future production levels. This would allow the EU to comply with current and future WTO requirements and dairy producers to capture most of the benefits of compensation. The budgetary constraint implies that full compensation is unlikely. It is also perhaps not necessary, at least from an economic point of view, given the excess profits currently enjoyed by the larger and most efficient dairy producers, which justify a modulation of compensation payments in function of farm size and location. CONCLUDING COMMENTS EU dairy farmers and processors are not well placed to take advantage of expected world market growth. Internal prices are too high relative to world prices, and the quota is too large relative to unsubsidized domestic and foreign demand. A two-tiered export policy is unlikely because of its more than likely WTO incompatibility (Bouamra Mechemache and Rquillart 1999). A radical option would be to eliminate quotas and intervention price

support to liberate the EU dairy industry from current and future WTO commitments (at least on the export side). Our experiments show that such a policy is currently feasible only in the context of politically acceptable compensation and of a minimum of frontier protection. The extent of the competitiveness of EU dairy farmers is obviously the key issue in determining the consequences of any quota removal policy. Whatever the compensation package and more generally whatever the future EU dairy policy, considerable restructuring will occur with a decrease in production costs, an increase in the scale of operations and a decline in the number of dairy farmers. This has already been happening for many years. Relative to other dairy exporters, the EU benefits form a definitive advantage due to the size of its domestic market. The EU should not forego this market opportunity. REFERENCES
Bouamra Mechemache, Z. and V. Rquillart. 1999. Policy reform in the European Union dairy sector: Effects on markets and welfare. Canadian Journal of Agricultural Economics, this issue. CAP Monitor. 1999. London: Agra Europe Ltd. European Commission. 1997. Situation and Outlook: Dairy Sector. Directorate General VI CAP 2000 Working Documents, April. European Commission. March 1999a. Agriculture Council: Policy agreement on CAP reform. Directorate General VI Newsletter (special edition), March. European Commission. March 1999b. Berlin European Council: Agenda 2000, Conclusions of the Presidency. Directorate General VI Newsletter (10), March. FAPRIUMC. 1998. Analysis of European Dairy Policy Options, with Additional Focus on Broader Effects of Agenda 2000. Policy Working Paper 1-98. Columbia, MO: University of Missouri, Food and Agricultural Policy Research Institute. FAPRIUMC. 1999. Implications of the Berlin Accord for EU Agriculture. Report 07-99. Columbia, MO: University of Missouri, Food and Agricultural Policy Research Institute. Fuller, F., J. Beghin, S. Mohanty, J. Fabiosa, C. Fang and P. Kaus. 1999. The impact of the Berlin accord and European enlargement on dairy markets. Canadian Journal of Agricultural Economics, this issue.

FUTURE OF EUROPEAN UNION DAIRY POLICY

101

Gohin, A. 1998. Modlisation du complexe agroalimentaire franais dans un cadre dquilibre gnral. PhD thesis. Paris: University of Paris I. Gohin, A. and H. Guyomard. 1999. Agenda 2000, dairy policy instruments and farm milk price. INRAESR working paper presented at the EAAE IXth congress, Warsaw, 2428 August. Gohin, A., H. Guyomard and C. Le Moul. 1999. Tariff protection reduction and Common Agricultural Policy reform: Implications of changes in methods of import demand modeling in a computable general equilibrium framework. INRAESR working paper presented at the Second Annual Conference on Global Economic Analysis, Assens, 2022 June. Guyomard, H., X. Delache, X. Irz and L.-P. Mah. 1996. A microeconometric analysis of milk quota transfer: Application to French producers. Journal of Agricultural Economics 47 (2): 20623. Josling, T. and S. Tangermann. 1999. The WTO agreement on agriculture and the next negotiating round. European Review of Agricultural Economics 26 (3): 37188. Lattimore, R. and R. Amor. 1998. World Dairy Policy and New Zealand. Discussion Paper 49. Lincoln University, International Trade Policy Research Centre, Commerce Division.

OLeary, E. and W. Fingleton. 1998. The impact of policy alternatives on Irish dairy farms. In Proceedings of the 1998 Agri-Food Economics Conference, edited by E. Pitts. Dublin: Teagasc. Organisation for Economic Cooperation and Development. 1999. OECD Agricultural Outlook, 19992004. Paris: OECD. Shoven, J. B. and J. Whalley. 1984. Applied general equilibrium models of taxation and international trade: an introductory survey. Journal of Economic Literature 22: 100751. Thompson, W., P. Liapis and P. Sckokai. 1999. Alternative trade mechanisms in world dairy markets. Canadian Journal of Agricultural Economics, this issue. United States. Department of Agriculture. 1998. 1998 EU Dairy Report. Washington, DC: USDA, Foreign Agricultural Service, Office of Agricultural Affairs, U.S. Mission to the EU. United States. Department of Agriculture. 1998. Agriculture in the WTO: Situation and Outlook Series. WRS-98-4. Washington, DC: USDA, ERS, December.