52004DC0505

Rapport présenté par la Commission - Le financement de l'Union européenne - Rapport de la Commission sur le fonctionnement du système des ressources propres

INTRODUCTION

L'article 9 de la décision actuelle sur les ressources propres [1] appelle la Commission à entreprendre

[1] Décision (2000/597, CE, Euratom) du Conseil relative au système des ressources propres des Communautés européennes, JO L 253 du 7.10.2000, p. 42.

«avant le 1er janvier 2006, un réexamen général du système des ressources propres, accompagné, le cas échéant, de propositions appropriées, en tenant compte de tous les facteurs pertinents, y compris, les effets de l'élargissement sur le financement du budget, la possibilité de modifier la structure des ressources propres en créant de nouvelles ressources propres autonomes et la correction des déséquilibres budgétaires accordée au Royaume-Uni ainsi que la réduction accordée à l'Allemagne, à l'Autriche, aux Pays-Bas et à la Suède, visée à l'article 5, paragraphe 1».

En réponse à une demande du Parlement européen et en accord avec le Conseil, la Commission s'est engagée à présenter le réexamen susmentionné sur le fonctionnement du système de financement avant la fin de 2004.

Le 10 février 2004, la Commission a adopté sa communication [2] sur les perspectives financières 2007-2013. La communication a identifié deux éléments principaux du système actuel des ressources propres méritant une attention particulière: premièrement, la transparence insuffisante du système pour les citoyens de l'UE combinée à une autonomie financière limitée vis-à-vis des budgets nationaux; deuxièmement, la nécessité de réformer le mécanisme existant pour la correction des déséquilibres budgétaires négatifs.

[2] COM(2004) 101 final du 10 février 2004.

Conformément à ces orientations, le présent rapport fournit un aperçu du système actuel et de ses principaux inconvénients, propose un mécanisme de correction généralisé au titre d'ajustement à court terme afin de traiter la principale question en suspens et expose pour le long terme une structure de ressources propres qui serait efficace, transparente et démocratique.

Une évaluation plus détaillée de ces points figure dans le document de travail de la Commission annexé au rapport.

1. le système des ressources propres actuel

Le système des ressources propres actuel est le résultat de modifications successives apportées au système original introduit en 1970. Le système comprend les trois catégories suivantes:

(1) Les ressources propres traditionnelles (RPT). Il s'agit principalement des droits de douane prélevés par les États membres au nom de l'UE.

(2) La ressource basée sur la taxe sur la valeur ajoutée (TVA). La ressource TVA est prélevée sur les bases TVA statistiques «fictives» et harmonisées des États membres, qui sont calculées sur la base des recettes TVA nationales.

En outre, la base TVA fictive est «écrêtée», le cas échéant, à hauteur de 50 % du revenu national brut de chaque État membre (RNB) pour réduire l'effet du caractère légèrement «régressif» de la TVA. Dans la pratique, cela a pour effet de transformer la ressource basée sur la TVA en ressource basée sur le RNB pour les pays concernés par la règle de l'écrêtement.

Le taux d'appel de la TVA ne peut pas dépasser 0,5 % de la base.

(3) La ressource basée sur le RNB. Il s'agit de la ressource résiduelle utilisée pour équilibrer le budget. Elle est prélevée à un taux uniforme proportionnellement au RNB de chaque État membre. La seule limite à ce taux est le plafond des ressources propres qui fixe le montant total de toutes les ressources propres à un maximum équivalent à 1,24 % du RNB de l'UE.

Enfin, un mécanisme spécifique destiné à corriger le déséquilibre budgétaire du Royaume-Uni fait également partie du système des ressources propres (voir section 3). Il a fait l'objet de nombreuses modifications, rendant le mécanisme de plus en plus complexe et opaque.

Les dernières modifications au système des ressources propres ont été arrêtées par le Conseil européen à Berlin en 1999. Ces modifications ont accentué la tendance au retrait des ressources propres traditionnelles et basées sur la TVA, tout en augmentant la part relative des contributions basées sur le PNB [3]/RNB.

[3] Ressource PNB (produit national brut) avant 2002.

Tableau 1 - La composition des ressources propres de l'UE

(en % du total des ressources propres; comptabilité de caisse)

>EMPLACEMENT TABLE>

1 À partir de 2002, le pourcentage des RPT conservé par les États membres au titre de compensation pour leurs frais de perception est passé de 10 % à 25 %. Cette différence a représenté environ 2,2 milliards d'euros en 2002, de même qu'en 2003.

2 Avant-projet de budget rectificatif 8/2004 (UE-25).

3 Avant-projet de budget 2005.

2. La performance du système des ressources propres actuel

2.1. Critères d'évaluation

Le système des ressources propres et les ressources propres individuelles peuvent être évalués au moyen de critères spécifiques. Les ressources propres individuelles ne sauraient répondre à tous les critères possibles d'évaluation. Néanmoins, un système fondé sur une combinaison de ressources de nature différente peut raisonnablement satisfaire aux principaux critères pertinents.

Les sept critères d'évaluation suivants ont été retenus dans le présent rapport:

* Visibilité et simplicité

* Autonomie financière

* Contribution à une meilleure affectation des ressources économiques

* Suffisance

* Bon rapport efficacité/coût au niveau administratif

* Stabilité des recettes

* Équité des contributions brutes

2.2. Évaluation

Le système actuel est performant au regard des critères de la suffisance et de la stabilité, mais il est clair qu'il ne remplit pas le critère de visibilité et de simplicité et qu'il ne contribue pas de manière significative à une répartition plus efficace des ressources économiques dans l'UE. Le système de financement est devenu de plus en plus complexe avec le temps, au point que même le citoyen intéressé éprouve des difficultés à comprendre son fonctionnement. L'autonomie financière est en outre de plus en plus limitée. Bien que le financement du budget soit assuré par des règles qui sont contraignantes pour tous les États membres, il n'y a pratiquement aucun lien direct avec les citoyens ou les contribuables. Le financement du budget repose plutôt sur des transferts effectués à partir des budgets nationaux. Quant aux autres critères, le système de financement actuel a révélé diverses lacunes, comme il est expliqué plus en détail dans le document de travail de la Commission annexé au présent rapport.

3. les soldes budgétaires négatifs excessifs

3.1. La question des déséquilibres budgétaires

L'Union européenne est une communauté de solidarité; certaines parties du budget de l'UE ont un objectif clair de solidarité tandis que d'autres financent la réalisation d'objectifs spécifiques dans le cadre de programmes communautaires convenus. Il existe donc des bénéficiaires nets du budget de l'UE et des contributeurs nets à ce budget, même si les avantages de cette politique profitent à l'Union dans son ensemble.

Il apparaît que les soldes budgétaires nets, qui se traduisent par les différences entre les contributions apportées au budget et les recettes obtenues, ne rendent pas pleinement compte du bénéfice résultant de l'appartenance à l'UE.

Toutefois, l'ampleur de certains de ces déséquilibres est au centre des discussions politiques. En 1984, le Conseil européen de Fontainebleau a introduit le mécanisme de correction existant au regard de la contribution nette d'un État membre, le RU. Ce mécanisme rembourse à ce pays les 2/3 de ses contributions nettes selon la notion des dépenses réparties. L'introduction de ce mécanisme a pour base le principe général que

«... tout État membre supportant une charge budgétaire excessive au regard de sa prospérité relative est susceptible de bénéficier, le moment venu, d'une correction»

Bien que la correction n'ait été décidée que pour le RU, le principe d'une correction généralisée a donc déjà été reconnu à Fontainebleau («tout État membre»), la correction devant reposer sur l'ampleur du déséquilibre budgétaire («excessive») et la richesse d'un État membre par comparaison avec l'ensemble de l'UE («prospérité relative»).

Au cours des 20 dernières années, les développements économiques, l'élargissement et les modifications apportées à la structure du budget de l'UE ont sensiblement altéré le contexte dans lequel le mécanisme existant de la correction britannique fonctionne.

Le tableau ci-dessous illustre l'évolution du revenu national brut (RNB) par habitant, exprimé en standard de pouvoir d'achat (SPA) [4], entre les années 1984 et 2003 pour tous les États membres qui ont été des contributeurs nets au budget de l'UE en 2002 [5].

[4] Le SPA est une monnaie artificielle qui reflète les différences dans les prix nationaux qui ne sont pas prises en compte par les taux de change. Cette unité permet des comparaisons de volume utiles d'indicateurs économiques entre les pays.

[5] Les soldes nets présentés dans ce document, tels que ceux utilisés pour la correction britannique, sont calculés avec inclusion des dépenses administratives. Pour cette raison, la Belgique et le Luxembourg n'apparaissent pas comme contributeurs nets. Comme la correction est payée un an plus tard, la correction 2007 sera versée en 2008.

>EMPLACEMENT TABLE>

En raison du changement radical de la position britannique par rapport aux autres contributeurs nets et en application du principe de Fontainebleau, l'octroi d'une correction sur une base exclusive au RU devrait être réexaminé. Par conséquent, le mécanisme existant devrait être transformé en un mécanisme de correction généralisé, répondant à l'objectif double consistant à

( prévenir les soldes budgétaires négatifs excessifs, en combinaison avec une réduction des différences entre les contributeurs nets à des niveaux comparables de prospérité;

( veiller à ce que les coûts de financement du mécanisme soient maintenus à un niveau raisonnable.

3.2. Impact du mécanisme de correction existant sur les soldes nets des États membres

Dans le cadre du mécanisme existant, le remboursement de la contribution nette du RU est de 66 %. Le mécanisme est financé par tous les États membres selon leurs parts respectives dans le RNB avec les exceptions suivantes:

- Le RU ne participe pas au financement de sa propre correction.

- La part de DE, NL, AT et SE est limitée à 25 % de leurs parts normales.

Tandis que la position nette britannique avant la correction n'a guère changé depuis son introduction (-0,48 % du RNB en 1985 et -0,47 % entre 1996 et 2002), sa position nette finale s'est améliorée sensiblement par rapport aux autres États membres à des niveaux similaires, voire inférieurs de prospérité, comme NL et DE (-0,47 % et -0,44 % respectivement).

Un facteur très important pour évaluer le mécanisme existant de la correction est l'impact de l'élargissement. Avec l'adhésion de nouveaux bénéficiaires nets, les soldes nets de tous les anciens États membres sont amenés à se dégrader car une part croissante des dépenses sera attribuée aux nouveaux États membres. Néanmoins, en raison du mécanisme actuel, le volume de la correction augmentera très sensiblement pour le RU. Le coût pour les autres États membres, y compris tous les nouveaux États membres, augmentera. Si le mécanisme de correction actuel reste en vigueur, la correction britannique augmenterait, selon les estimations de la Commission et sur la base de sa proposition pour les perspectives financières 2007-2013, de plus de 50 %.

3.3. Le mécanisme de correction généralisé proposé

La Commission propose une généralisation du mécanisme de correction afin de rapprocher le système de l'objectif initial consistant à éviter les charges budgétaires excessives. En introduisant une sorte de «filet de sécurité» pour les grands contributeurs nets dont la contribution nette dépasse un certain niveau, cela pourrait également faciliter une approche plus constructive pour garantir que les moyens budgétaires répondent aux défis politiques de l'Union européenne élargie.

La future correction doit être calculée sur la base du solde budgétaire net de chaque État membre en relation avec le budget de l'UE. Le mécanisme devrait être déclenché si les contributions nettes dépassent un seuil, exprimé en pourcentage du RNB de chaque État membre, reflétant le niveau minimal de solidarité financière illimitée autorisé entre les États membres. Les positions nettes supérieures à ce seuil pourront faire l'objet d'une correction (remboursement partielle), offrant ainsi une garantie contre les contributions nettes excessives. Inversement, le volume total des corrections (volume de remboursement) sera limité à un montant maximal, offrant à ceux qui ne bénéficient pas d'une correction une garantie contre les coûts excessifs du mécanisme. Si la somme de toutes les corrections excède le volume total prédéterminé, le taux de remboursement sera réduit en conséquence.

Tout mécanisme de correction implique des complications pour le financement du budget. C'est pourquoi la proposition de base doit être relativement simple et suffisamment transparente. À cet égard, les paramètres actuels utilisés pour le calcul de la correction britannique ne doivent être modifiés que si cela s'avère nécessaire et ils doivent être simplifiés chaque fois que possible.

Le nouveau mécanisme proposé par la Commission comporte les éléments suivants:

- établissement d'un niveau seuil comme pourcentage du RNB;

- écrêtement du volume total des corrections;

- simplification du financement des corrections sur la base des parts RNB, tous les États membres participant au financement du montant global des corrections proportionnellement à leur prospérité relative;

- maintien de la notion de dépenses réparties et du taux de remboursement jusqu'au volume maximal.

Selon cette proposition, le RU serait de loin le plus grand bénéficiaire du mécanisme de correction généralisé. D'après la proposition de la Commission sur les prochaines perspectives financières et la proposition sur un mécanisme de correction généralisé, le RU recevrait en moyenne une compensation nette environ deux fois plus élevée que celle revenant au Danemark.

>EMPLACEMENT TABLE>

Le tableau ci-dessus donne une estimation des soldes budgétaires nets moyens pour les États membres résultant du mécanisme de correction généralisé proposé pour la période s'étendant jusque 2013 et compare les chiffres avec la continuation du mécanisme actuel et une situation exempte de correction.

Le mécanisme de correction généralisé diminuera les soldes nets négatifs, réduira l'écart entre les contributeurs nets, et, par ailleurs, amoindrira la charge de financement de ceux qui ne bénéficient pas du mécanisme [6].

[6] Voir l'annexe technique pour plus de détails.

La Commission présente séparément une proposition spécifique visant à modifier la décision actuelle sur les ressources propres en vue d'introduire le mécanisme de correction généralisé proposé.

4. La structure des ressources propres

4.1. La structure des ressources propres optimale

Trois alternatives de base pour le financement du budget de l'UE sont examinées ci-dessous. Dans chacun des trois cas, les ressources propres traditionnelles (RPT) seraient maintenues car la perception des droits d'importation constitue un instrument de la politique commerciale de l'Union dont le produit revient «naturellement» à l'Union.

(1) Maintien du système de financement actuel

Le système actuel a assuré un financement sans heurt du budget de l'UE. Cependant, dans sa forme actuelle, le système de financement est dépourvu de lien direct avec les citoyens. Les ajustements effectués dans le passé pour satisfaire des intérêts spécifiques ont accentué l'opacité du système. Le grand public ne perçoit donc pas les conséquences budgétaires des politiques de l'Union. Avec le poids énorme de la ressource RNB, les États membres, et notamment les contributeurs nets, tendent à juger les politiques et les initiatives de l'UE exclusivement au regard de leur imputation nationale et sans faire grand cas du contenu des politiques, avec le risque d'obscurcir la valeur ajoutée des politiques communautaires.

En raison de l'absence de tout lien, y compris informatif, avec les citoyens de l'UE et de l'incitation croissante à se concentrer sur la notion budgétaire étriquée du «juste retour», le système actuel doit être réformé.

(2) Un système de financement basé exclusivement sur le RNB

Dans le cadre d'un système de financement exclusivement basé sur la clé RNB, l'Union dépendrait entièrement des «contributions» des États membres. L'avantage de ce système est qu'il serait simple et facile à comprendre. Les «contributions» seraient étroitement liées à la prospérité relative des États membres. Mais assurer le financement du budget par les contributions des États membres peut être approprié pour une organisation internationale telle que les Nations Unies. Un tel système ne reflète cependant pas le statut de l'Union européenne. Cette approche impliquerait l'idée d'une Union dans laquelle les citoyens ne seraient représentés qu'indirectement par leurs États membres. Le statut de l'UE en tant qu'Union des États membres et des citoyens, qui est actuellement reflétée dans le traité, serait abandonné du côté du financement du budget, entraînant le renoncement à la notion de ressources «propres», qui a constitué une pierre angulaire des finances de l'UE depuis la première décision sur les ressources propres en 1970. La Commission rejette une telle alternative.

(3) Un système de financement basé sur des ressources propres fiscales

Un système basé dans une large mesure sur des ressources propres fiscales a le potentiel d'augmenter l'autonomie financière du budget de l'UE en créant des relations financières plus directes entre le budget et les citoyens de l'UE. Le fait que les individus et les opérateurs économiques deviennent des contributeurs et la réduction correspondante des contributions des États membres au financement du budget de l'UE se solderaient par une plus grande visibilité pour les citoyens et une responsabilité politique accrue de l'autorité budgétaire quant aux décisions de dépenses. Cette approche permettrait aussi de ne plus centrer le débat politique sur l'aspect étriqué des contributions nationales mais de l'orienter vers le contenu des politiques communautaires et l'intérêt européen général. Enfin, une visibilité accrue renforcerait la confiance dans les finances de l'UE, dont le volume inférieur à 2,5 % des dépenses publiques totales dans l'UE est souvent surestimé.

Toutefois, même un financement entièrement fiscal du budget de l'UE ne semble pas approprié. Comme l'exigence d'un budget équilibré fixée par l'article 268 du traité n'est pas et ne devrait pas être remise en question, une ressource fiscale assumant le rôle actuel de RNB comme ressource d'équilibrage «résiduelle» impliquerait un taux de taxation variable selon les besoins budgétaires ou nécessiterait une plus grande flexibilité du côté des dépenses que ce qui est autorisé par le cadre actuel des perspectives financières. Cependant, les modifications fréquentes du taux de taxation seraient en pratique mal acceptées par les contribuables (incertitude juridique, procédures techniquement et administrativement lourdes) et pourraient entrer en conflit avec les fiscalités nationales. Une approche progressive, consistant à maintenir une ressource RNB limitée tout en augmentant la part des ressources fiscales, semble préférable.

Une réforme de la structure des ressources propres devrait donc être axée sur l'introduction neutre du point de vue budgétaire d'une nouvelle ressource propre fiscale représentant jusqu'à la moitié du budget.

5. Vers un système de financement basé sur les contributions des États membres et des citoyens

5.1. Augmentation de la part des ressources propres fiscales

La réforme du système de financement devrait remédier aux inconvénients du système actuel, c'est-à-dire l'absence de lien direct avec les citoyens de l'UE, la dépendance à l'égard des transferts effectués à partir des budgets nationaux et une complexité injustifiée, tout en contribuant dans la mesure du possible à une attribution efficace des ressources. Les autres critères deviennent moins pertinents pour les nouvelles ressources, à mesure que le système global leur permet d'être remplis à un degré raisonnable.

Pour des raisons d'équité entre les citoyens dans les différents États membres, l'introduction de nouvelles ressources fiscales requiert une harmonisation préalable suffisante de l'assiette. Le choix de la ressource fiscale la plus appropriée dépend de manière cruciale du degré d'harmonisation effectif de l'assiette.

L'existence de l'arbitraire régional [7] ou de facteurs externes transfrontaliers plaide très nettement en faveur de l'harmonisation des assiettes et des taux, ce qui pourrait aussi justifier l'affectation de tout ou partie des recettes fiscales correspondantes au niveau de l'UE. Cela pourrait contribuer à réduire la répartition inefficace des ressources économiques.

[7] L'arbitraire régional concerne une situation dans laquelle il est difficile de déterminer la part exacte d'une assiette à allouer aux différents États membres ou s'il existe un écart élevé (potentiel) entre le pays percevant la taxe et le pays de résidence des agents économiques supportant la charge fiscale.

L'augmentation de la part des ressources propres fiscales pour le budget de l'UE ne requiert pas de nouvelles taxes. Il faut néanmoins prendre une décision quant au partage des recettes ou des taux de taxation entre les niveaux national et communautaire. La part de l'UE pourrait être prélevée dans le cadre du taux national payé par les assujettis. Le budget total de l'UE, de toute façon limité par le plafond des ressources propres à un montant maximum équivalent à 1,24 % du RNB de l'UE, n'augmenterait pas car les recettes provenant de la ressource fiscale seraient compensées par une diminution correspondante de la ressource RNB actuelle.

Les sections suivantes illustrent trois options principales visant à améliorer le fonctionnement du système de financement de l'UE. Toutes les trois maintiennent la ressource RNB actuelle au titre de ressource d'équilibrage résiduel ainsi que les ressources propres traditionnelles, mais supposent que jusqu'à la moitié du budget est financée par une nouvelle ressource fiscale qui remplacerait également la TVA « statistique » actuelle.

5.2. Option 1 - Un système des ressources propres doté de ressources fiscales liées à la consommation d'énergie

En vertu de la nouvelle directive sur la taxation de l'énergie [8], la plupart des produits d'énergie sont soumis à des taxes communautaires. Bien que la directive adoptée sur la taxation de l'énergie n'ait pas été préparée aux fins d'introduire de nouvelles ressources propres, elle crée néanmoins des conditions qui lui sont favorables en harmonisant les assiettes et en fixant des taux minimaux.

[8] Directive 2003/96/CE du Conseil du 27 octobre 2003 restructurant le cadre communautaire de taxation des produits énergétiques et de l'électricité (JO L 283 du 31.10.2003).

Une ressource fiscale sur les produits d'énergie ne devrait pas toutefois être basée sur tous les produits couverts par la directive. Un prélèvement communautaire pourrait être limité à l'assiette liée au carburant utilisé pour les transports routiers (essence plombée et sans plomb, diesel, LPG et gaz naturel utilisé pour le transport). Cela serait relativement simple d'un point de vue administratif puisque l'assiette est déjà harmonisée au niveau communautaire et que la plupart des possibilités de différenciation fiscale autorisées par la directive s'appliquent à d'autres produits d'énergie. Des taux communautaires inférieurs à la moitié des taux minimaux prévus par la directive sur la taxation de l'énergie seraient suffisants pour financer la moitié du budget de l'UE. Les décisions principales influeraient sur les taux à appliquer et les variations possibles selon le produit et l'utilisation.

Un prélèvement communautaire sur le carburant d'aviation ou les émissions connexes pourrait compléter utilement un prélèvement sur le carburant destiné aux transports routiers. Le système européen des transports aériens est hautement intégré et les émissions d'avions dépassent les frontières nationales. De bons arguments plaident donc en faveur d'une taxation des émissions au niveau de l'UE afin d'internaliser les coûts socio-économiques externes du changement climatique et des autres effets sur l'environnement causés par l'aviation dans le prix des transports aériens. Néanmoins, le carburant d'aviation (kérosène) est actuellement exonéré pour les vols transfrontaliers.

D'un point de vue technique, l'introduction d'une ressource fiscale basée sur le carburant utilisé pour les transports routiers serait possible à relativement brève échéance (environ 3 à 6 ans).

5.3. Option 2 - Un système de ressources propres doté d'une ressource fiscale TVA

En raison de son lien direct avec la consommation quotidienne et contrairement à la ressource TVA «statistique» actuelle, l'application d'un taux communautaire aux bases TVA nationales créerait un lien direct clair entre le financement du budget de l'UE et les citoyens et rendrait ces derniers plus conscients des coûts de l'Union.

Une ressource TVA véritablement fiscale serait mise en oeuvre par le biais d'un taux communautaire dans le cadre du taux TVA national payé par les assujettis. Le taux serait prélevé avec le taux national sur la même base d'imposition. Les citoyens n'auraient pas à supporter de charge fiscale supplémentaire puisque le taux communautaire serait compensé par une diminution équivalente du taux de TVA national. Par exemple, si le taux national était de 21 %, et le taux communautaire de 1 %, le taux national tomberait à 20 %. Le taux total de TVA prélevé serait toujours de 21 %.

À des fins de visibilité, la TVA communautaire et la TVA nationale devraient figurer séparément sur les factures ou les relevés fournis par l'assujetti.

Un taux communautaire de 1 % serait suffisant pour couvrir environ la moitié des besoins de financement du budget de l'UE.

Les difficultés potentielles de cette proposition sont dues à l'harmonisation incomplète des régimes TVA des États membres et sont principalement liées aux marchandises taxées au taux zéro. Une autre difficulté potentielle pourrait provenir d'une part différente de la base TVA dans le revenu national selon les États membres (le « caractère régressif » de la TVA), qui incite depuis 1988 les États membres à réduire sa part dans les ressources propres. D'autre part, du point de vue du consommateur, le taux communautaire aurait le même impact sur des consommateurs comparables à travers l'Union [9].

[9] En supposant que l'effet du taux zéro soit neutralisé (cf. annexe à ce rapport).

Techniquement, l'introduction d'un taux TVA communautaire serait possible à relativement brève échéance (jusqu'à 6 ans).

5.4. Option 3 - Un système des ressources propres doté d'une ressource basée sur le revenu des sociétés

Quant aux deux autres options, une ressource fiscale communautaire basée sur le revenu des sociétés exigerait aussi la définition préalable d'une assiette (consolidée) commune qui serait appliquée aux sociétés. Cette harmonisation contribuerait au bon fonctionnement du marché intérieur et à une répartition plus efficace des ressources économiques qui est actuellement faussée par des facteurs externes transfrontaliers. Aujourd'hui, l'existence de 25 régimes fiscaux nationaux et la multiplicité des lois, conventions et pratiques fiscales représentent en soi un obstacle à l'activité économique transfrontalière. Elle impose des coûts d'adaptation considérables aux sociétés opérant au-delà des frontières et génère de nombreuses niches fiscales.

Le travail actuel sur une réforme complète de la fiscalité des entreprises se concentre sur la notion d'une assiette (consolidée) commune, selon toutes probabilités pour un sous-groupe d'États membres intéressés, dans le but de supprimer les entraves fiscales au marché intérieur. Ce travail n'envisage aucune action sur les taux de taxation et il n'est pas non plus conçu comme une méthode permettant de collecter des recettes pour le budget de l'UE.

L'option d'un impôt sur les sociétés supposerait la fixation d'un taux de taxation minimal à l'assiette harmonisée.

Les recettes provenant des impôts sur les sociétés sont importantes, avec un total dans l'UE représentant actuellement en moyenne 2,6 % du RNB total de l'UE. Ainsi, pour les besoins du budget de l'UE, moins d'un quart de ces recettes devrait être attribué à l'UE.

Compte tenu de la situation actuelle, cette alternative serait la plus longue à mettre en oeuvre, tant du point de vue politique qu'administratif.

6. Résumé et conclusions

L'introduction d'une nouvelle ressource propre fiscale remplaçant la ressource « statistique » actuelle basée sur la TVA et finançant une partie significative du budget de l'UE permettrait de remédier aux lacunes principales du système actuel, c'est-à-dire l'absence de lien direct avec les citoyens de l'UE, une dépendance quasi-exclusive à l'égard des transferts effectués à partir des budgets nationaux et une complexité injustifiée. Elle pourrait également contribuer à une meilleure attribution des ressources économiques dans l'UE. Même en représentant une part plus faible dans le total des ressources propres que dans le cadre du système actuel, la ressource basée sur le RNB continuerait de jouer un rôle important et fournirait la garantie que le système satisfait raisonnablement à tous les critères pertinents.

La Commission propose trois candidats principaux pour les éventuelles ressources propres fiscales: une ressource basée sur 1. la consommation d'énergie, 2. les bases TVA nationales et 3. le revenu d'entreprise.

Une ressource basée sur la consommation d'énergie et conçue comme un prélèvement communautaire sur le carburant destiné aux transports routiers serait une source suffisante et stable de financement pour le budget de l'UE et créerait un lien direct avec les citoyens. L'assiette est déjà harmonisée au niveau de l'UE. Elle pourrait être assortie d'un prélèvement communautaire sur le carburant d'aviation ou les émissions connexes, pour mettre ainsi un terme à l'exonération actuelle du carburéacteur et fixer un prix pour les coûts environnementaux de l'aviation.

L'harmonisation de l'assiette dans le domaine de la TVA est assez avancée et elle est une source suffisante et stable de recettes. Une ressource TVA fiscale rendrait le financement de l'UE hautement transparent pour les citoyens de l'UE. Elle serait également évolutionnaire puisqu'elle impliquerait une réforme des dispositions existantes plutôt que l'introduction d'une ressource totalement nouvelle. D'un point de vue administratif, son introduction ne présenterait aucune difficulté insurmontable.

Dans le contexte d'une politique commune et en raison de facteurs externes transfrontaliers, les recettes provenant d'une assiette harmonisée de l'impôt sur les sociétés constitueraient également une source de financement appropriée pour le budget de l'UE.

L'Union européenne est une union d'États membres et de citoyens. Chacune des trois options examinées ci-dessus transposerait cette notion dans le domaine du financement du budget de l'UE. Le renforcement du lien direct des citoyens avec le budget permettrait aussi de focaliser les débats sur les dépenses sur le contenu plutôt que sur les «positions nettes» du budget «national».

Évidemment, toute nouvelle attribution de ressource au budget de l'UE doit non seulement être décidée à l'unanimité par le Conseil, mais elle doit aussi être ratifiée par les Parlements de tous les États membres.

La mise en oeuvre d'une ressource basée sur l'énergie ou sur la TVA serait possible dans le moyen terme, tandis qu'une ressource fiscale basée sur le revenu d'entreprise doit être considérée comme une option à beaucoup plus long terme. Il importe maintenant de disposer d'une orientation politique visant à préparer les conditions pour la réforme de la structure des ressources propres existantes. À cet égard, la Commission invite le Conseil à:

- examiner les options proposées dans le présent rapport;

- prendre note de l'intention de la Commission de préparer une feuille de route en vue de remplacer, sur la base d'une proposition de la Commission, la ressource TVA actuelle par une véritable ressource propre fiscale d'ici à 2014.

Afin d'apporter une solution à court terme à la question des déséquilibres budgétaires excessifs, qui soit opérationnelle dès le début des prochaines perspectives financières, la Commission propose

- d'introduire un mécanisme de correction généralisé afin de rectifier les déséquilibres budgétaires excessifs conformément à la proposition de décision du Conseil sur le système des ressources propres et les mesures d'accompagnement proposées.

TECHNICAL ANNEX

Financing the European Union Commission

report on the operation of the own resources system

INTRODUCTION

Article 9 of the current own resources decision [1] calls on the Commission to

[1] Council decision (2000/597, EC, Euratom) on the system of the European Communities' own resources, OJ L 253, 7.10.2000, p. 42.

'undertake, before 1 January 2006, a general review of the own resources system, accompanied, if necessary, by appropriate proposals, in the light of all relevant factors, including the effects of enlargement on the financing of the budget, the possibility of modifying the structure of the own resources by creating new autonomous own resources and the correction of budgetary imbalances granted to the United Kingdom as well as the granting to Austria, Germany, the Netherlands and Sweden of the reduction pursuant to Article 5(1)'.

In response to a request from the European Parliament and in agreement with the Council, the Commission committed itself to present the abovementioned review on the functioning of the financing system before the end of 2004.

On 10 February 2004, the Commission adopted its communication [2] on 'Building our common future - Policy challenges and budgetary means of the enlarged Union 2007-2013'. The communication identified two main elements of the current own resources system deserving closer attention: first, the insufficient transparency of the system for EU citizens combined with limited financial autonomy from national treasuries; secondly, the need to reform the existing mechanism for correction of negative budgetary imbalances.

[2] COM(2004) 101 final of 10 February 2004.

This report reviews those issues in more detail. For this purpose it is divided in four parts.

Part I of the report presents the main features of the current system, the relevant assessment criteria and provides an assessment of the functioning of the current financing system.

Part II examines the existing mechanism to correct budgetary imbalances and proposes to replace it with a generalised mechanism for correcting excessive negative budgetary balances in order to ensure an equitable treatment of net contributors at comparable levels of prosperity and financing costs that are kept at a reasonable level.

Part III reviews alternative scenarios for the system of financing the EU budget and outlines possible own resources systems that in the longer term might allow certain drawbacks of the current own resources system to be overcome. It is important to stress that future modifications of the own resources system in line with what is proposed in Part III would probably require a review of the generalised correction mechanism proposed in Part II.

Thus, whereas part II is to be seen as the short term adjustment of the current system in order to address the main outstanding issue, part III offers for the longer term the outline of a system that would be more effective, transparent and democratic.

Finally, Part IV of the report presents the final comments.

PART I - THE PERFORMANCE OF THE CURRENT OWN RESOURCES SYSTEM

This part provides an overview of the current system of own resources, presents the criteria used for assessing the performance of own resources individually and together as a system, and then provides an overall assessment.

1. The current own resources system

The current own resources system is the result of successive modifications of the original system introduced in 1970. The latest modifications were decided by the European Council in Berlin in 1999 and implemented through the own resources decision of 29 September 2000 [3] (see below).

[3] See footnote 1.

The current own resources may be divided into three categories that are used to finance the budget in a sequential way, i.e. recourse is made to the following category only when the previous one is exhausted. In practice this means that the third category, the one related to the gross national income (GNI) of Member States, is the residual one used to balance the budget. It is also the only resource to be affected by modifications of budgeted expenditure during the implementation of the budget, i.e. stemming from amendments to the budget. The three categories of own resources are the following:

(1) So-called traditional own resources (TOR). These are mainly [4] customs duties and are collected by Member States on behalf of the EU. Member States retain a fixed percentage of the amounts collected as a compensation for their costs of collecting them. The percentage was increased from 10 % to 25 % as from 2001.

[4] Traditional own resources also include agricultural duties, i.e. import duties collected on agricultural products, and sugar levies. Together, these two categories are estimated to account for around 10 % of total TOR and 1.2 % of total own resources in 2004).

(2) The resource based on value added tax (VAT). This resource is levied on the notional harmonised VAT bases of Member States. The statistical 'notional' VAT bases are calculated in order to compensate for differences in national VAT regimes due to incomplete harmonisation of VAT at EU level. The notional VAT base is calculated, for each Member State, by dividing total national VAT receipts by the so-called weighted average rate of VAT. The weighted average rate is derived from macro-economic statistics (mainly national accounts). In order to take account of the specific national procedures and arrive at a harmonised base for all Member States, changes are made, either to the net revenue collected (known as 'corrections'), or to the VAT base (known as 'financial compensations'). This implies, once again, that statistical elements are used in the calculation of the VAT base rather than fiscal data.

Furthermore, the notional VAT base is 'capped', where applicable [5], at 50 % of each Member States' GNI to reduce the effect of the 'regressive' character [6] of VAT. In practice, this turns the VAT-based own resource into a GNI-based resource for the countries concerned by the capping rule. Any increase in VAT receipts in a Member State whose VAT base is capped will have no effect on the EU VAT resource.

[5] Forecasts for 2005 indicate that for 13 out of 25 Member States their VAT base will exceed 50 % of their GNI.

[6] The share of consumption in national income tends to be higher in relatively less prosperous countries. As a result, VAT revenue tends to be proportionally higher than in more prosperous countries.

A uniform percentage rate is levied on the capped and harmonised VAT bases of all Member States. This call rate cannot exceed 0.5 % of the base but is furthermore reduced to take into account the theoretical impact of the amount of the UK correction (see below).

(3) The GNI-based resource. This resource is levied as a uniform rate in proportion to the GNI of each Member State. There is no particular limit on this rate, other than the own resources ceiling that limits the total amount of all own resources to a maximum of 1.24 % of the EU's GNI.

Finally, a specific mechanism for correcting the budgetary imbalance of the United Kingdom is also part of the own resources system. The basic principle of the correction mechanism is to reimburse the UK by 66 % of its budgetary imbalance. The mechanism reduces the own resources payments of the UK and increases the payments of all other Member States, including the 10 new Member States joining in 2004. The correction mechanism has been modified on several occasions since its introduction in 1985 in order to neutralise the impact on the UK budgetary balance of each subsequent modification of the own resources system. Several additional layers have therefore been added to the original calculation, rendering the mechanism increasingly complex and non-transparent.

According to the rules of the own resources system the amount of the UK correction also has an impact on the rate of call of VAT, which is reduced by a percentage theoretically needed to finance the correction ('the frozen rate'). This is a relic from the time the correction was financed on the basis of Member States' shares in the EU VAT base, whereas since 1988 the financing is in reality calculated on the basis of GNI.

The cost of the correction is borne by the other Member States in proportion to their GNI. However, in order to alleviate the budgetary imbalances of four other Member States (Austria, Germany, the Netherlands and Sweden) these countries pay only 25 % of their normal share. The financing of this reduction is added to the payments of the remaining 20 Member States.

The latest changes to the own resources system were implemented through the current own resources decision of 29 September 2000 and notably concerned the following elements:

- An increase in the percentage of traditional own resources retained by Member States from 10 % to 25 %.

- A reduction of the maximum rate of call of the VAT resource from 1 % to 0.75 % in 2002 and 0.5 % in 2004.

- Some changes to the method of calculating the UK correction in order to compensate for effects related to enlargement and the increase in the percentage of traditional own resources retained by Member States.

- A reduction of the share of Austria, Germany, the Netherlands and Sweden in the financing of the UK correction to one fourth of its normal value. Previously only Germany had a reduction, by one third, of its financing share.

- The European system of accounts 1995 (ESA 95) replaced the previous version, ESA 79, also in the budgetary and own resources area. As a consequence gross national income (GNI) replaced the concept of gross national product (GNP) in the area of the EU budget.

This package further reinforced the declining trend of traditional and VAT-based own resources and the corresponding increase in the relative share of the GNP/GNI- based contributions.

Table 1 - The composition of EU own resources

(in per cent of total own resources; cash basis)

>EMPLACEMENT TABLE>

1 As from 2002 the % of TOR retained by Member States as a compensation for their collection costs

was raised from 10 % to 25 %. This difference represented about EUR 2.2 billion in 2002 as well as in 2003.

2 Preliminary draft amending budget 8/2004 (EU-25).

3 Preliminary draft budget 2005.

2. Assessment criteria

The own resources system and individual own resources can be assessed against specific criteria. It is virtually impossible for individual own resources to satisfy all criteria. However, a system based on a combination of resources of different natures may reasonably meet the different criteria. The 'systemic' assessment is discussed in more detail in part III of the report.

The following seven assessment criteria are relevant to this report:

- Visibility and simplicity; i.e. the financing of the EU budget should be visible to and understood by EU citizens.

- Financial autonomy; i.e. the financing of the EU budget should provide a sufficient degree of autonomy from national treasuries to reduce the tendency towards a narrow focus on national interest. Autonomy is enhanced if the financing relies to a significant degree on a direct link with citizens/taxpayers.

- Efficient allocation; i.e. the financing of the EU should contribute to an efficient allocation of economic resources, for example by contributing to the proper functioning of the internal market or by setting a price tag on negative externalities, such as pollution.

- Sufficiency; i.e. the resources used for financing the EU must be sufficient to cover the expenditure of the EU in the long run. As for individual own resources, their yield should be significant in relation to the size of the EU budget.

- Cost-effectiveness; the administration costs should be low relative to the yield.

- Stability; the resources should be reasonably stable over time.

- Equity in gross contributions; i.e. the burden should be fairly shared among Member States as well as citizens.

A more extensive presentation of these criteria can be found in annex I to this report.

3. Assessing the performance of the current own resources system

To summarise the conclusions of this section, the current system has performed well as regards the criteria of sufficiency and stability, but clearly fails to fulfil the visibility and simplicity criterion and it does not contribute significantly to a more efficient allocation of economic resources. The financing system has grown increasingly complex over time, making it difficult even for the interested citizen to understand how it works. Financial autonomy is, furthermore, becoming increasingly limited. Although the financing of the budget is ensured by rules that are binding for all Member States, there is virtually no direct link to citizens or tax-payers. Instead the financing of the budget relies on transfers from national treasuries. As for the remaining criteria the current financing system has shown various shortcomings, as explained more in detail below.

3.1. Visibility and simplicity

The current financing system is mainly and increasingly based on national contributions paid directly by the treasuries of Member States to the Commission and lacks a visible direct link to citizens. Most citizens do not know what they pay to the EU budget. This gives much scope for misperceptions of the costs and for a feeling of unfair burden-sharing.

In the current system, traditional own resources (TOR) is the only direct link between the EU budget and the citizens, since they are levied directly on the economic agents paying the import duty. However, since it is a relatively small number of agents that bear the direct burden of the tax and since TOR only represent a very small part of total own resources, they do not contribute much to the visibility of the EU budget.

The lack of a direct and visible link to the EU citizens also results in reduced accountability of the European Parliament, which might make it less sensitive to the cost of financing different policy measures.

Furthermore, the current system has grown increasingly complex over time, making it difficult even for the interested citizen to understand how it works. The VAT resource is levied on a theoretical harmonised base in order to compensate for national derogations from Community VAT legislation and varying tax rates, requiring complicated calculations. Moreover, because of the tendency of the VAT base to be relatively higher in less prosperous Member States (i.e. the so-called regressivity of the VAT base), rules for limiting the base as a percentage of the national income ('capping') have been introduced. The VAT-based contribution from capped countries is therefore in practice based on their gross national income. The main reason for the lack of transparency of the system is, however, the UK correction. Not only is the correction mechanism itself highly complex, the amount of the correction also influences the rate of call of both the VAT and the GNI resource. Finally, in order to alleviate the net budgetary position of other large net contributors, some Member States have been granted a 'rebate on the rebate', i.e. a reduction on the financing of the correction, the cost of which is borne by the remaining Member States.

3.2. Financial autonomy

The rules governing the financing of the budget are laid down in detail in the own resources decision and its different implementing provisions, which are binding for all Member States. The financing of expenditure is therefore guaranteed up to the own resources ceiling. This also means that the financing of the annual budget is completely technical with no room for political discussions. The budgetary authority's room of manoeuvre is extremely restricted, since the level and composition of revenue are largely determined by the own resources decision and by the strict application of the rule of budgetary balance. The financial sufficiency of the EU budget is ensured. As for financial autonomy from the treasuries of the Member States the picture is, however, quite different.

Traditional own resources (TOR) currently represent the only truly fiscal own resources of the EU. They are collected by Member States on behalf of the EU, after deduction of 25 % of the amount collected. The percentage deducted represents an increase compared to the 10 % retained by Member States prior to 2001 (see also next section). As a result, the weight of these fiscal resources in the financing of the EU budget was reduced. TOR do, however, only represent around 10 % of total own resources.

The other own resources are calculated as a percentage of statistical aggregates (GNI or the notional harmonised VAT base) and paid directly by the treasuries of Member States to the Commission. The GNI-resource, initially envisaged as the 'residual' resource to ensure a balanced budget, has in fact become the overwhelming source of revenue for the EU budget.

The increasing dependence of the EU budget on inter-governmental transfers from national treasuries, accounting for almost 90 % of EU total revenue in later years, encourages Member States to seek to maximise ill-defined concepts of national benefit from the EU budget. Such a trend may well impair the future EU budget to adequately reflect the new challenges and needs to meet the legitimate expectations and concerns of all European citizens. In shaping a European Union with more than 25 Member States, narrow national interests will need to be pushed into the background, allowing instead for serious and more focused discussions about common European concerns. Increased autonomy of the EU budget from national treasuries would contribute to a more constructive approach to the benefits of the Union's policies.

Financing the budget by contributions of the members is adequate for an international organisation. It does not necessarily reflect the status of the European Union and is in contradiction with the spirit of the treaty provision, introduced in 1970, whereby the Community budget should be financed by own resources.

3.3. Efficient allocation of economic resources

As the current own resources are mainly levied directly on the treasuries of the Member States, they do not have a direct impact on the relative prices in the economy and the behaviour of the economic agents.

It is only the traditional own resources (TOR) that have a direct impact on relative prices (on imported goods relative to goods produced in the Union) and therefore on the allocation decisions of the economic agents. TOR are, however, the result of common policies such as the customs union and the common agricultural policy and were not primarily introduced for the purpose of financing the EU budget. Due to the mismatch between the country collecting the duties and the country of residence of the economic agents bearing the economic burden of this tax it was then logical to attribute the corresponding revenue to the EU level. The introduction of a Common Customs Tariff has contributed to an efficient allocation of economic resources between Member States by creating a level playing field for import of goods into the Union.

In this context it could also be mentioned that the increase from 10 % to 25 % in the percentage of traditional own resources retained by Member States was introduced irrespective of the actual collection costs. To a large extent it was the result of difficult negotiations about ways to correct excessive negative budgetary balances. The current level could, however, be seen as a useful incentive for Member States to collect TOR. On the other hand, it is not necessarily the Member States bearing the highest costs related to the protection of the external border of the Union that collect the duties and retain the corresponding 25 % of the collected amount. This is because customs operations are undertaken in stages in different Member States. Under the present system, Member States undertaking controls at the external frontier may be in a less favourable position than those Member States actually collecting the duties. In this regard, it is not surprising that some Member States have concluded bilateral agreements, especially in the context of simplified customs procedures, in order to share the 25 % compensation for the costs of collecting import duties in order to better reflect the actual collection and control costs.

Although the present VAT-based resource has today in practice evolved into a national contribution, it earlier served as an impetus for establishing a common VAT system and therefore contributed to the proper functioning of the internal market.

3.4. Sufficiency and stability

The present system has generated sufficient and stable resources to finance expenditure plans. The decline of traditional and VAT-based own resources is automatically compensated by the GNI-based resource, which guarantees financing up to the own resources ceiling of 1.24 % of total EU GNI. The margin available under the own resources ceiling in the current financial perspective is around 0.15 % of EU GNI for the whole 2000-2006 period.

3.5. Cost-effectiveness

The collection of TOR and VAT is the duty of Member States. The collection of traditional own resources is cumbersome and demands significant administrative resources. However, import duties are primarily levied to protect producers in the Union [7] and not to finance the EU budget. The costs of collecting them should therefore only marginally be charged to the financing rules of the EU. Furthermore, if import duties were not to be attributed to the EU level, given the mismatch between the country collecting the duties and the country of residence of the economic agents bearing the economic burden of this tax, complex tax-sharing rules would normally have to be defined, which would entail additional costs.

[7] Furthermore, in shaping its common commercial policy the Union does not act alone but in the multilateral context of the World Trade Organisation.

The VAT-based resource imposes an additional workload on national administrations, since the harmonised VAT resources base is only calculated for the purposes of the EU budget. However, this can be considered as acceptable in relation to its potential yield. The administrative cost of the GNI resource is marginal, since it is incurred for the establishment of national accounts, irrespective of the EU budget.

3.6. Equity

Equity may refer to the situation of individuals or to the fair burden-sharing among Member States. Since the budget is currently financed almost exclusively by contributions from Member States, the assessment here focuses on equity between Member States.

As the ability to contribute to the EU budget can best be measured by a nation's national income, the increasing importance of the GNI [8] resource means that equity between Member States in the financing of the budget has improved over time. As the GNI resource now accounts for ¾ of total own resources in the 2004 budget, Member States' contributions are, generally speaking, becoming closely correlated to income levels across Member States.

[8] Prior to 2002 the GNP (gross national product) resource.

There is, however, one important exception to this rule. Due to the specific correction mechanism in its favour, the UK contributes clearly less than the other Member States in relation to its GNI. Furthermore, the UK correction also distorts equity indirectly since the financing of the correction by the other Member States is not proportional to GNI. The distorting effect of the UK correction is, furthermore expected to increase in the future due to enlargements of the Union.

PART II - PREVENTING EXCESSIVE NEGATIVE BUDGETARY IMBALANCES

1. Introduction

The European Union creates a community of solidarity with parts of the EU budget serving a clear cohesion goal, whereas other parts fund the achievement of specific objectives through EU agreed programmes. As a result, net beneficiaries of, and net contributors to, the EU budget will always exist, although the policy benefits accrue to the Union as a whole.

Budgetary balances (also called net balances), measured by the difference between contributions to and receipts from the EU budget, fail to account fully for the benefits resulting from EU membership. For example, research or border protection expenditure benefits not only the immediate recipients but also gives rise to spill-over effects transcending national borders [9]. It may not be possible to quantify the extent of these spill-overs, but their consideration would modify the assessment of the accounting imbalances.

[9] These spill-over effects include the spending of income generated in the receiving Member State on goods or services produced in another Member State, the purchase of financial asset denominated in various Member States currencies etc.

The definition of budgetary balances is also fraught with significant conceptual and accounting problems. Numerous choices have to be made in computing budgetary balances on the items to be included in the receipts and expenditure flows, and on the reference periods (e.g. cash vs. accrual figures, surpluses from previous years, etc.). The resulting budgetary balances vary significantly depending on the choices made.

Nevertheless, the size of some of these imbalances has been at the centre of discussions. After years of budgetary stalemate and agonizing discussions in 1984 the Fontainebleau European Council introduced the existing UK correction, which was then given effect by the own resources decision of 7 May 1985. The decision was based on the following general principle [10]:

[10] See Fontainebleau European Council, 'Conclusions of the Presidency', Bulletin of the European Communities, 6-1984.

'... any Member State sustaining a budgetary burden which is excessive in relation to its relative prosperity may benefit from a correction at the appropriate time.'

The principle of a generalised correction was therefore already acknowledged by the European Council in 1984 ('any Member State'). The decision for granting a correction should be based on two criteria: the size of the budgetary imbalance ('excessive') and the wealth of a Member State compared to the EU as a whole ('relative prosperity').

The justifications for granting a correction on an exclusive basis to the UK are today less justified than at the time of the Fontainebleau European Council, since several other Member States can legitimately claim that their current situation is comparable to the UK's. Furthermore, the cost of enlargement should be fairly shared. Consequently, it is argued that, in light of the principle set in the Fontainebleau conclusions, the conditions exist for the introduction of a generalised correction mechanism. The parameters defining such a correction mechanism are examined and a proposal is made with the twin goals of:

- preventing excessive negative budgetary balances combined with a reduction of differences between net contributors at comparable levels of prosperity;

- ensuring that the financing costs of the mechanism are kept at a reasonable level.

After briefly recalling the origin and main effects of the existing UK correction (Section 2), this report provides a brief analysis of the UK's relative prosperity and net budgetary balance in comparison with the other net contributors (Section 3). The subsequent sections of the report contain an analysis of the likely evolution of the UK correction till 2013 (Section 4) and examine proposals for a possible phasing out of that correction in its current form (Section 5). Section 6 defines the key parameters for a generalised correction mechanism and, finally, Section 7 proposes a new generalised mechanism to correct excessive negative budgetary imbalances.

2. The budgetary compensation for the United Kingdom

2.1. Origin of the mechanism

The budgetary imbalance of the United Kingdom was at the centre of the political debate for about a decade (1974-1984), frequently provoking stalemates in the EU decision-making process. The question appeared immediately after the accession of the country to the European Community. The large negative budgetary balance of the UK at that time was essentially due to two factors:

(a) an agricultural sector relatively smaller and structurally different from those of other Member States, which results in lower CAP spending in the United Kingdom;

(b) a proportionally larger contribution to the financing of the EU budget [11].

[11] The United Kingdom had a relatively higher share in the harmonised VAT base than in the total GNP of the EU. At that time the VAT resource was the primary source of EU revenue. The GNI resource was introduced in 1988.

Furthermore, it should be recalled that during the first years after its accession, the United Kingdom was one of the least prosperous Member States with a GDP per capita well below the EU average [12]. This tended to exacerbate the problem of the UK's negative budgetary balance.

[12] In the five years preceding the introduction of the UK correction (1979-1983), only Ireland (and as of 1981 Greece) had a lower GDP per capita than the UK, expressed in purchasing power standards (PPS).

The issue was finally settled at the Fontainebleau European Council of 1984 and the resulting rebate mechanism constitutes an integral part of the own resources decisions (ORD) taken subsequently. Being an integral part of the ORD, the UK rebate continues until modified or abolished by a new ORD. Modifications to the ORD require unanimity in Council and ratification by all national Parliaments. The rebate was reviewed in 1988, 1992 and 1999 and on all occasions the European Council decided to maintain it.

2.2. Description of the mechanism

While technically very complicated to calculate, the underlying principle of the UK correction mechanism is quite simple: to reimburse to the UK 66% of its net contribution to the EU budget. In essence this is calculated in three steps.

- The UK's percentage share in allocated expenditure (i.e. what the UK receives from the EU budget) is subtracted from the UK's percentage share in payments to the EU budget [13].

[13] These payments exclude payments of so-called traditional own resources, which are considered as resources that cannot be allocated to individual Member States.

- This difference is multiplied by the total amount of allocated expenditure [14] to obtain a measure of the UK's net contribution to the EU budget.

[14] Allocated expenditure includes all utilised appropriations for payments, with the exception of external expenditure (namely expenditure headings 4, 6 and 7 of the current financial perspective 2000-2006) and expenditure that cannot be reasonable allocated or identified. However, pre-accession expenditure relating to actual payments in the last year before accession of any acceding country is also deducted from allocated expenditure.

- The result is multiplied by 0.66 (i.e. 66% percent of the whole net contribution is reimbursed).

Most of the additional complications in the calculation arise from the fact that the overall contribution to the EU budget actually paid by the UK (including so-called traditional own resources and the UK correction) has to be equivalent to the overall contribution that the UK would have paid if the rules of the 1985 ORD (i.e. the rules that were applicable as a result of the Fontainebleau European Council) were still in force today. In practice this means that all subsequent modifications to the own resources system, introduced by successive ORD's, have to be neutralised through the calculation of the UK rebate. This includes the effect of the introduction of the GNP/GNI resource, the reduction of the VAT call rate, the capping of the VAT base and the increase of the traditional own resources collection costs [15].

[15] Besides the issues mentioned in this paragraph, the UK rebate also complicates the calculation of the VAT own resource through the mechanism of the so-called 'frozen' rate of call for VAT. While the original logic behind the 'frozen' rate disappeared in 1988 with the introduction of the GNP own resource, the mechanism continues to exist today. It now functions as a complicated and roundabout way to reduce the VAT own resource.

The financing of the correction for a given year 't' is made up by all the Member States in accordance with their respective percentage shares in GNI payments in the year t+1, with the following exceptions:

- the UK does not participate in the financing of its own rebate;

- the share of DE, NL, AT and SE is restricted to 25% of the shares resulting from the calculation above. The difference is made up by the remaining 10 (20 since May 2004) Member States according to their share in GNI payments.

3. Current situation of the UK compared to other net contributors

This section examines the relative prosperity and the size of the net budgetary balances for all net contributors to the EU budget.

3.1. Relative degree of prosperity

The table below gives an overview of the gross national income (GNI) per capita expressed in purchasing power standards (PPS [16]) for the years 2003 and 1984 for all Member States that were net contributor in the year 2002 [17]. (The 2003 figures for all 25 Member States are presented in the annexed Table I.

[16] The PPS is an artificial currency that reflects differences in national price levels that are not taken into account by exchange rates. This unit allows meaningful volume comparisons of economic indicators among countries. Data on PPS are calculated by Eurostat.

[17] Net balances presented in this document, like those used for the UK correction, are calculated including administrative expenditure. For this reason, Belgium and Luxembourg do not appear among the net contributors.

>EMPLACEMENT TABLE>

In 2003, GNI per capita of all net contributors to the EU budget, expressed in PPS, ranges between 97% and 111% of the EU-15 average. At 111.2%, the UK's relative prosperity is at the top of the range. This is in sharp contrast with the situation in 1984, when the UK was the least prosperous of the net contributors and only Greece and Ireland had a lower GNI per capita than the UK.

In view of the considerable shift in the UK's position compared to the other net contributors, it is therefore necessary to reassess whether the UK correction mechanism as it stands is still justified in light of the Fontainebleau principle whereby a Member State's net balance should be viewed in relation to its relative prosperity.

3.2. Net budgetary balances before UK correction

The net budgetary balance of the UK for the year 1985 (the first year for which the correction was calculated) amounted to -0.48% of GNI before correction. (Although accurate data are not available for the other Member States, Germany was the only other major net contributor to the EU budget at that time.) As illustrated in the table below, on average the UK's net balance has remained at a comparable level in recent years. Over the period 1996-2002 the net budgetary balance of EU-15 net contributors - before UK correction and including administrative expenditure - was on average the following:

Table 2

Net budgetary balances before UK correction

for EU-15 selected members (annual averages 1996-2002)

// in % of GNI

United Kingdom // -0.47%

Germany // -0.44%

Netherlands // -0.43%

Sweden // -0.38%

Austria // -0.24%

Italy // -0.06%

France // -0.04%

Average net budgetary balances for the period 2008-2013 [18] will deteriorate for all net contributors across-the-board because of the financing cost of enlargement. According to internal Commission estimates and assuming expenditure levels equal to the financial perspective ceilings as proposed by the Commission in its communication of 10 February 2003 [19], the estimated net balances of the net contributors would be the following. (The figures for all 25 Member States are presented in the annexed Table II).

[18] The net balances for the period 2007-2013 have been estimated on the basis of the expenditure level (payment appropriations) as proposed by the Commission for the EU-27. Where feasible, the allocation across Member States was based on a detailed calculation on the basis of an established methodology (e.g. cohesion expenditure). In other cases, an indicative breakdown was used on the basis of historical data. Expenditure for the 10 new Member States that acceded on 1 May 2004 was calculated according to an identical methodology as for the EU-15, except for the phasing-in of the EAGGF Guarantee direct payments. Expenditure for Bulgaria and Romania is included in a manner which is coherent with the Commission communication Financial package for the accession negotiations with Bulgaria and Romania, SEC(2004) 160/4 adopted on 10 February 2004.

[19] COM(2004) 101 final.

Table 3

Estimated net budgetary balances before UK correction

for EU-15 selected members (annual averages 2008-2013 [20])

[20] As corrections are reimbursed one year later, averages in this table and the following ones are calculated over a six-year period since the 2007 correction is paid in 2008 and the 2013 correction takes place in 2014.

// in % of GNI

United Kingdom // -0.62%

Netherlands // -0.55%

Germany // -0.52%

Sweden // -0.47%

Austria // -0.37%

Italy // -0.29%

France // -0.27%

Denmark // -0.20%

Finland // -0.14%

Under the assumption that the level of agricultural expenditure for the EU-25 agreed by the Brussels European Council in October 2002 and the 'cohesion' expenditure proposed by the Commission under the so-called Objective 1 and Cohesion fund remained unchanged, a reduction of the overall level of payment appropriations from the 1.14% of GNI proposed by the Commission to, say, 1.00%, would only have a very limited impact on the size of the estimated net balances. This is because the expenditure for the new Member States would in essence not be touched and the reduction could only be achieved by drastically cutting the other non-agricultural expenditure going to the EU-15 and/or external expenditure that does not enter in the calculation of net balances. As a consequence, the resulting reduction of own resources payments for the net contributors would to a large extent be offset by a corresponding reduction in EU allocated expenditure in these same Member States.

In the absence of any correction mechanism, the UK would have been on average the largest net contributor over the last 7 years, and would probably remain the largest net contributor to the EU budget over the period 2007-2013. However, the net balances of NL, DE and, to a lesser extent, SE have been and are expected to remain of a comparable order of magnitude. As shown in the previous section, all these three Member States are currently relatively less prosperous than the UK.

Whether the size of net budgetary balances is 'excessive' in view of the relative prosperity of the Member States concerned largely depends on the political perception of the acceptable degree of financial solidarity within the Union. If the UK net balance was judged 'excessive', then the application of the Fontainebleau principle would rather point to the extension of a correction mechanism to other Member States, who - at lower levels of prosperity - bear negative net balances of a similar order of magnitude.

In any event, granting a correction on an exclusive basis to only one Member State appears to be unjustified, especially when taking into account the expected evolution of net budgetary balances in the enlarged Union under unchanged conditions, as illustrated in the next section.

4. Estimated net balances with unchanged Own Resources Decision

Estimates indicate that over the period 2007-2013 the UK correction will increase by more than 50% compared to the average over the latest 7 years to reach an estimated EUR 7.1 billion from EUR 4.6 billion in the period 1997-2003.

Although the envisaged enlargement to 10 new Member States was unanimously agreed at the Berlin European Council in March 1999, the UK insisted and obtained that enlargement-related expenditure be taken into account [21] when calculating the UK correction, thus shielding it from most of the financial consequences of enlargement. That is the main reason for the expected future increase in the UK correction.

[21] Pre-accession expenditure relating to actual payments in the last year before accession of any acceding country is permanently deducted from the allocated expenditure.

As a consequence, the cost for the Member States that pay a full share in the financing of the UK correction, including all the new Member States, will increase proportionally. The average estimated financing cost over the period 2008-2013 for all 25 Member States is presented in the annexed Table IV.

Should the current own resources decision remain in force, the average net balance for the net contributors over the 2007-2013 period after UK correction is estimated as follows (Table III in annex II provides estimates for all 25 Member States):

Table 4

Estimated net budgetary balances after UK correction

for all expected net contributors

(annual averages 2008-2013)

// in % of GNI

Netherlands // -0.56%

Germany // -0.54%

Sweden // -0.50%

Italy // -0.41%

Austria // -0.38%

Cyprus [22] // -0.37%

[22] Throughout this document estimates are based on areas controlled by the Republic of Cyprus.

France // -0.37%

Denmark // -0.31%

Finland // -0.25%

United Kingdom // -0.25%

According to these estimates, the UK will become (together with Finland) the smallest net contributor to the EU budget, therefore clearly departing from the Fontainebleau principle mentioned above. The persistence of such a growing anomaly risks undermining the legitimacy of EU policies in general, particularly in new Member States. The current system of a unique correction for the UK therefore cannot continue. Instead, the generalization of the correction mechanism, evolving from the existing correction would allow bringing the system closer to the original objective of avoiding excessive budgetary burdens in relation to the relative prosperity of Member States. By introducing a sort of 'safety net' for large net contributors beyond a certain level, it may also facilitate a more constructive approach to ensure the budgetary means necessary to meet the policy challenges of the enlarged Union.

5. Reduction or phasing-out of the UK correction

5.1. Estimated net balances with a reduction of the refund rate

One simple way to rectify the expected 'overshooting' of the UK correction would be to reduce the reimbursement percentage from 66% to - for instance - 33%.

The results in terms of net budgetary balances, expressed as a percentage of each Member State's GNI are summarised below for the net contributors:

Table 5

Estimated net budgetary balances after UK correction

with refund rate reduced to 33%

for all expected net contributors

(annual averages 2008-2013)

// in % of GNI

Netherlands // -0.55%

Germany // -0.53%

Sweden // -0.48%

United Kingdom // -0.46%

Austria // -0.38%

Italy // -0.34%

Cyprus [23] // -0.32%

[23] Estimates based on areas controlled by the Republic of Cyprus.

France // -0.32%

Denmark // -0.25%

Finland // -0.19%

The UK's net contribution would be more in line with that of other net contributors, although it would remain rather low in comparison with some of the other net contributors that have lower levels of relative prosperity. Among these net contributors, F and I would benefit most from a reduction of the rebate since these countries unlike AT, DE, NL and SE do not benefit from a reduction in their financing share of the UK correction.

Conversely, the net position of NL, DE and SE hardly improves. A reduction of the UK rebate would therefore not address the perceived excessive net balances of these Member States. Furthermore, it would perpetuate the situation whereby one Member State continues to receive a correction on an exclusive basis, while others in an almost identical, or relatively worse, situation do not.

5.2. Estimated net balances with the UK correction frozen at its current

level

An alternative to the reduction of the refund rate could be to freeze the amount of the correction at a certain level. For instance, the UK correction could be frozen at the level actually enjoyed by the UK on average over the period 1996-2002, which is around EUR 4.5 billion. However, as the result would be comparable to a reduction of the refund rate, the same disadvantages would persist.

Table 6

Estimated net budgetary balances after UK correction

with UK correction frozen at its current level

for all expected net contributors

(annual averages 2008-2013)

// in % of GNI

Netherlands // -0.55%

Germany // -0.53%

Sweden // -0.49%

United Kingdom // -0.38%

Austria // -0.38%

Italy // -0.37%

Cyprus [24] // -0.34%

[24] Estimates based on areas controlled by the Republic of Cyprus.

France // -0.34%

Denmark // -0.27%

Finland // -0.21%

5.3. Estimated net balances with a phasing-out of the UK rebate

Phasing-out the UK rebate over the period covered by the next financial perspective, as advocated by certain Member States, would certainly be the easiest solution to limit the unintended effects of the existing UK correction. In its simplest form, this could be done by leaving the calculation method unchanged, while progressively reducing the result by a given percentage over a certain number of years, say over seven years. Thus, the 2007 rebate (to be paid and financed in 2008) could amount to 6/7th of the amount obtained according to the current calculation method, the 2008 rebate to 5/7th and so on. The 2012 rebate, to be paid and financed in 2013, would amount to 1/7th of the normal calculation. In this example it would also constitute the last correction.

The estimated net balances of the net contributors over the period 2007-2013 resulting from such a phasing-out regime are presented below:

Table 7

Estimated net budgetary balances after UK correction

with a linear phasing-out of the UK correction over 7 years

for all expected net contributors

(annual averages 2008-2013)

// in % of GNI

Netherlands // -0.55%

Germany // -0.53%

Sweden // -0.49%

United Kingdom // -0.43%

Austria // -0.38%

Italy // -0.35%

Cyprus [25] // -0.32%

[25] Estimates based on areas controlled by the Republic of Cyprus.

France // -0.32%

Denmark // -0.26%

Finland // -0.19%

However, it should be borne in mind that the figures in the table above represent the average for the phasing-out period. After the period under consideration, the UK would no longer benefit from any correction mechanism. The phasing-out of UK correction would have only a minor impact on the net balances of the other large net contributors, since they are to a large extent protected from the financing cost of the UK rebate. Therefore, this alternative would not solve the problem of the 'excessive' net balances of the other net contributors.

5.4. Estimated net balances with a phasing-out of the frozen UK

correction

Alternatively, as advocated by certain Member States, the UK correction could be frozen at an agreed level (for instance EUR 4.5 billion, equivalent to the average over the period 1996-2002) and then phased-out linearly over a number of years. This is the equivalent of combining the effects discussed in Sections 5.2 and 5.3 above. The disadvantages mentioned in the previous paragraphs would remain unchanged.

Table 8

Estimated net budgetary balances after UK correction

with a linear phasing-out of the frozen UK correction over 7 years

for all expected net contributors

(annual averages 2008-2013)

// in % of GNI

Netherlands // -0.55%

Germany // -0.53%

United Kingdom // -0.50%

Sweden // -0.48%

Austria // -0.37%

Italy // -0.33%

Cyprus [26] // -0.31%

[26] Estimates based on areas controlled by the Republic of Cyprus.

France // -0.31%

Denmark // -0.23%

Finland // -0.17%

5.5. Estimated net balances with enlargement-related expenditure

excluded from the calculation of the UK correction

The calculation of the UK correction takes into account all the expenditure that is 'internal' to the EU and that can reasonably be allocated to an individual Member State [27]. Expenditure in the new Member States is also taken into account, with the exception of a small correction related to the amount of the pre-accession expenditure in the year before accession. The expenditure in the new Member States is the principal cause of the expected 50% increase in the size of the UK correction. It is therefore appropriate to consider the effect of continuing the current system while fully excluding the expenditure in the new Member States from the calculation.

[27] This issue is discussed under Section 6.2.

This approach would have the advantage of at least assuring fair burden sharing between all Member States of the cost related to the current and the future enlargements of the EU.

If the expenditure going to the 12 new Member States would be excluded, the UK correction is expected to be on average reduced to EUR 4.2 billion over the period 2007-2013. The results in terms of net budgetary balances, expressed as a percentage of each Member State's GNI are summarised below for the net contributors:

Table 9

Estimated net budgetary balances after UK correction

with enlargement-related expenditure excluded

for all expected net contributors

(annual averages 2008-2013)

// in % of GNI

Netherlands // -0.55%

Germany // -0.53%

Sweden // -0.49%

United Kingdom // -0.39%

Austria // -0.38%

Italy // -0.37%

Cyprus // -0.33%

France // -0.33%

Denmark // -0.27%

Finland // -0.21%

The net balances of DE, AT, SE and NL would only be affected in a very marginal way compared to the situation under the unchanged own resources decision, since these countries pay only 25% of their normal financing share of the financing of the UK correction.

Again, such a solution would not solve the problem whereby only 1 Member State would be eligible for a correction, while others at comparable or lower levels of relative prosperity and bearing negative net balances of similar size would continue to be excluded. Furthermore, it would effectively create in a durable way two categories of EU expenditure (within the EU-15 and in the new Member States).

6. Parameters of a generalised correction mechanism

6.1. Introduction

The proposed generalised correction mechanism has to be seen in the context of the Commission overall 'package' for the post-2006 financial framework. The final outcome is thus likely to be affected by decisions to be taken on the expenditure side of the package.

If it is accepted that the net contributions of certain Member States are excessive, then none of the alternatives presented in Sections 3 through 5 namely, unchanged own resources decision, abolishing, reducing or phasing-out the UK correction - seems to generate acceptable results in terms of equitable burden sharing among Member States. That is why the Commission proposes to introduce a generalised corrective mechanism.

The generalised correction mechanism is a tool for correcting 'excessive' net budgetary balances in an own resources system primarily financed through contributions from the Member State's treasuries. Any substantial changes to the own resources system (see Part III of this report), would of course require a re-assessment of the need for a corrective mechanism.

A generalised correction mechanism would be calculated on the basis of the net budgetary balance of each Member State in relation to the budget of the EU. The mechanism should be triggered beyond a threshold, expressed as a percentage of each Member State's GNI, reflecting the minimum accepted level of unlimited financial solidarity between Member States and representing a sort of basic reasonable net contribution. Net positions exceeding such a threshold will be eligible for a correction (partial refund). The amount of the correction is to be based on the part of the net balance exceeding this threshold, multiplied by a refund rate (i.e. the percentage of the amount in excess of the agreed threshold to be compensated). The amount corresponding to the sum of all corrections could be capped at a total predetermined volume.

The net balance would be calculated on the basis of a zero-sum situation, i.e. the sum of the net balances of all Member States should be equal to zero [28]. (This principle is also applied in the calculation of the UK correction.)

[28] For a discussion of the different types of net balances and their calculation methods, see the Commission's 1998 own resources report, Annex 3.

6.2. Categories of expenditure to be taken into account

The current definition of allocated expenditure used in the calculation of the UK rebate as well as the Commission's yearly report on the allocation of expenditure to Member States is based on the principle whereby all expenditure for which it can reasonably be shown that the recipient is a resident of a certain Member State is being allocated to that Member State. Recipients are not limited to public bodies but also include private firms, individuals, etc...The allocation is limited to the first round effect [29]. The Commission recognises that this instrument is an imperfect tool and that certain problems of correct allocation continue to exist [30].

[29] This means the Commission does not include possible second-round effects into the allocation. Often cited examples of these effects are for instance the possible repatriation of profits by foreign firms or consortia involved in cohesion projects and salaries of European officials residing in Belgium or Luxembourg that is not necessarily exclusively spent in these countries.

[30] Although the Commission is constantly trying to refine the data to improve the quality of the allocation, a number of problems continue to persist. Agricultural export subsidies for instance are allocated to the country exporting the goods, an allocation method which tends to unduly increase the expenditure allocated to gateway countries such as Belgium and the Netherlands. Likewise it is not always easy to identify exactly the final beneficiary of certain payments made to international consortia, for example for research expenditure. Since these consortia often have their administrative base located near the EU institutions, an exaggerated share in this type of expenditure may be allocated to Belgium and Luxembourg.

According to this definition, allocated expenditure includes all categories of expenditure that are 'internal' to the EU. This includes virtually the total amount of expenditure headings 1 (agriculture) and 2 (structural operations) as well as the vast majority of expenditure under headings 3 (internal policies) and 5 (administration), excluding only some minor items of expenditure that cannot be reasonably allocated to any individual Member State [31]. Headings 4 (external policies), 6 (reserves) and 7 (pre-accession expenditure) relate explicitly to 'external' expenditure and are therefore not taken into account when calculating the net budgetary balance of a Member State.

[31] In terms of the newly proposed expenditure headings in the Commission communication of 10 February 2003, this corresponds roughly with the content of the new headings 1a, 1b, 2, 3 and administrative expenditure.

It has on occasion been argued that certain categories of this internal expenditure should be fully or partially excluded from the calculation of the net balances. The arguments advanced to defend this are based on perceived conceptual and practical difficulties related to expenditure allocation. However, it should be noted that, while the European Council has reviewed the calculation method of the UK correction on several occasions, it has each time decided to keep the expenditure categories included in the calculation unchanged. Indeed, excluding any of these expenditure categories would constitute a different approach than the basic principle evoked at the beginning of this section. That is why the Commission proposes to keep the current accepted definition unchanged [32]. Thus, all expenditure that can reasonably be allocated to an individual Member State (i.e. broadly speaking, the headings 1, 2, 3 and 5 under the 2000-2006 financial perspective) would continue to be taken into account for the calculation of the net budgetary balance in the context of the proposed generalised correction mechanism.

[32] The only envisageable amendment concerns the agricultural export refunds which could be excluded on the grounds that there is no clear link between the port of exit and the final beneficiary of these interventions. A similar argument is used to exclude traditional own resources from the revenue to be taken into account (cf. section 6.3).

6.3. Categories of revenue to be taken into account

The calculation of the UK rebate excludes the traditional own resources (TOR) from the categories of revenue to be taken into account. This is justified for 2 reasons. First, TOR are not considered to be a Member State contribution but constitute revenue belonging to the EU by virtue of the customs union. Second, TOR are often levied at the port of entry into the EU in application of the Common Customs Code. Since the final consumer of the imported goods does not necessarily live in the country of the port of entry there is no clear geographical link between the collection of the duties and the economic burden they represent. TOR should therefore continue to be excluded in any future proposal to introduce a generalised correction mechanism.

Consequently, the combined share in VAT and GNI own resources would be used as the proper base to calculate any Member States' share in EU budgetary revenue.

A proposal for a generalised correction mechanism should be as simple and transparent as possible. Therefore all existing complications in the UK rebate calculation which are related to the preservation of the UK's 1984 revenue position (share in uncapped VAT, calculating the 'advantage' as well as the 'TOR windfall gains'), should be abolished.

6.4. The level of the threshold

6.4.1. Fixed threshold

Simulations have been run for different levels of the threshold (using a refund rate of 66%) ranging from zero to 0.50% of GNI. The results are summarised in the annexed Tables V through IX [33]:

[33] A minus (-) sign means an increase of a Member State's total own resources payments compared to the estimated future (2008-2013 average) situation under an unchanged own resources decision. Conversely a plus (+) sign means a reduction in total own resources payments for the Member State concerned.

- Table V: estimates of the level of the gross corrections;

- Table VI: estimates of the financing of the corrections;

- Table VII: estimates of the net level of the corrections (= sum of tables V + VI);

- Table VIII: estimates of the impact of introducing a GCM compared to the situation with unchanged own resources decision;

- Table IX: estimates of the net budgetary balances as a percentage of GNI;

The total sum of the corrections at different threshold levels is estimated as follows (details for individual Member States are presented in Table V):

Table 10

Estimated gross corrections

(average 2008-2013)

Level of the threshold

(as a % of GNI, EU-27) // Sum of all corrections

(in billion of euros)

0.00% // 25.8

0.10% // 19.8

0.20% // 13.8

0.25% // 11.1

0.30% // 8.8

0.40% // 5.2

0.50% // 1.9

A threshold level of around -0.25% would represent a sort of neutral point whereby the estimated future cost of financing the generalised correction mechanism would be equal to the estimated future cost of financing the current UK correction mechanism. This is illustrated in Table V in annex II: the effect of introducing a GCM is zero or close to zero at a threshold of 0.25% for all Member States that are currently financing a full share in the UK correction [34]. A threshold lower than 0.25% would mean that the Member States paying in full the UK rebate (they currently bear more than 90% of the total cost) would have to pay more than the estimated cost of financing the current UK correction mechanism in the future. With thresholds higher than 0.25% the mechanism would become less costly for those Member States, including all net recipients from the EU budget, than financing the estimated future UK correction mechanism.

[34] This is valid for all Member States that do not receive a gross correction and therefore only participate in the financing. However, as can be observed in Table V in the annex, at a threshold of 0.25% a few of the 20 Member States that currently pay a full share in the financing of the UK correction would actually receive a gross correction. Consequently, as illustrated in Table VIII, these Member States would still be better off at 0.25% compared to the future cost of financing the UK correction.

The table above (as well as the annexed Table V) illustrates that the total sum of the corrections with a 0.25% threshold would be substantially higher than the estimated level of the future UK correction with the current own resources decision (ORD) unchanged. Applying a threshold of -0.25% would generate a level of (gross) corrections slightly above EUR 11 billion, which is significantly higher than the approximately EUR 7 billion average (net) UK correction for the same period. This difference is linked to the assumption that the financing of the corrections would be shared by all Member States according to their share in GNI (see section 6.7 below).

Thus, for those Member States who would not benefit from the GCM, an overall volume of up to EUR 11 billion of corrections could be financed before the GCM became more expensive than the continuation of the current ORD.

The impact of introducing a generalised mechanism with a threshold is not linear. The effect on Member States' net balances, compared to the current situation, is influenced by a combination of the following three elements: a) the level of the Member States' budgetary imbalance before correction; b) the level of the threshold; and c) the financing rules under the existing UK correction.

The estimated impact is shown in the annexed Table VIII on the basis of which the following observations can be made:

- Net beneficiaries from the budget do not receive any corrections, regardless of the threshold level. As previously indicated, the cost of financing the correction mechanism becomes more expensive for these countries compared to the future financing cost of the UK correction at threshold levels lower than 0.25%.

- All major net contributors to the EU budget (UK, DE, NL, SE) would benefit from the correction mechanism with thresholds lower than or equal to 0.50% of GNI.

- Smaller net contributors would receive corrections depending on the threshold level. The relative position of each individual net contributor, compared to their estimated future position under the current ORD, is determined by a combination of the following two factors:

- the size of the negative net balance before correction, whereby larger net contributors such as DE, NL, SE tend to benefit more at lower threshold levels (compared to smaller net contributors);

- the current financing regime of the UK correction, whereby Member States currently benefiting from a special arrangement (DE, NL, SE, AT) tend to benefit less at higher threshold levels (compared to smaller net contributors).

- The case of AT merits a specific mention since it illustrates the mechanism at work here. Since this Member State is expected to be a smaller net contributor than NL, DE and SE, it would receive smaller corrections (comparable to the corrections received by FR and IT). However, as it would lose the 75% reduction in the financing of the UK correction (whereas countries such as FR and IT do not enjoy such a reduction at present) it would see its net balance deteriorate at higher threshold levels (compared to the estimated situation with unchanged own resources decision).

- The UK would experience a deterioration of its net balance at any threshold level, even in the situation where there is no threshold at all. In this case the UK would receive a correction of an order of magnitude similar to the current UK correction but it would have to participate in the financing of both its own and the other Member States' corrections. It is worth recalling that the deterioration of the UK net balance illustrated in the tables above is in comparison with the UK's estimated future position, i.e. in comparison with the situation where the UK is expected to become the smallest net contributor to the EU budget.

Any future 'fairer' mechanism should ensure in particular that the resulting financing cost does not lead to a heavier burden than under the current UK correction mechanism for 'cohesion' Member States.

6.4.2. Variable threshold

It is possible to include an element of progressivity into the correction mechanism by introducing a variable threshold related to the relative prosperity of each Member State rather than the fixed threshold illustrated in the previous paragraphs. This would ensure a closer link between levels of relative prosperity and net balances. In other words, relatively more prosperous Member States would be expected to have proportionally larger negative budgetary imbalances than less prosperous Member States.

One possible approach would be to use a multi-layer threshold, based on each Member States' GNI per capita expressed in purchasing power standards (PPS). By way of example, the results of a simulation based on the following differentiation criteria are presented below:

GNI per capita PPS (where EU-15 avg. = 100) // Threshold

(in % of GNI)

110% < PPS // - 0.40%

100% < PPS <110% // - 0.35%

PPS < 100% // - 0.30%

The variable threshold, based on the multi-layer schedule given above, would be the following for the net contributors to the budget:

Member State // Threshold

Denmark // - 0.40%

Germany // -0.30%

France // -0.35%

Italy // -0.30%

Cyprus // -0.30%

Netherlands // -0.35%

Austria // -0.35%

Finland // -0.35%

Sweden // -0.35%

UK // -0.40%

The UK's threshold would be increased to -0.40% while the threshold applied to DE, IT and CY would be reduced to -0.30%. The threshold for the other net contributors would not be affected.

Using such a variable rather than a fixed threshold would only have a limited impact on the gross level of the corrections, since the reduction of the correction for certain Member states would of necessity be (partially) offset by the increased correction for others. Fundamentally, any introduction of progressivity in the correction mechanism will -ceteris paribus- benefit DE (and possibly IT) while deteriorating the position of the UK. Extending or decreasing the range of the threshold variability will not modify this.

6.5. Maximum Available Refund Volume

A legitimate concern of many net beneficiaries of the EU budget is that the financing cost of a generalised correction mechanism could spiral out of control at some point. Thus, if the financing cost exceeded the future financing cost of the UK correction, the generalised correction mechanism would become more expensive to finance than the unchanged UK correction. As a safeguard against such a risk, a maximum available refund volume (MARV), i.e. a cap, could be introduced. This cap would be applied automatically to the total level of corrections in case the sum of these corrections exceeds an agreed MARV. All corrections would then be linearly reduced. Such a cap (MARV) would be determined ex-ante as an ad hoc amount valid for the duration of the financial perspective.

Fixing the level of the MARV is clearly a discretionary decision.

In any case, it seems reasonable to assume that the cost of financing the corrections should not exceed the equivalent of the estimated future financing cost of the UK correction for those Member States currently financing a full share (i.e. all except UK, DE, NL, SE and AT).

6.6. Percentage of excess negative balance to be corrected

Beyond the agreed threshold, a certain percentage of the net budgetary balance is to be corrected. The percentage used in the existing UK correction is equal to 66%. This variable is purely linear. For instance, at a given threshold a refund of 66% of the excess balance will yield a correction which is exactly twice as big as a correction with a 33% refund percentage. Since the European Council of Fontainebleau agreed on a refund rate of 66%, and given the relatively low sensitivity of results to different refund rates, it is proposed to keep this parameter at the same level used for the current UK correction mechanism.

Additionally, the Commission proposes as a safeguard for the net beneficiaries who might be concerned about the future evolution of the financing cost of the correction mechanism to make the refund rate the adjustment variable with 66% being the maximum rate. This maximum rate would normally be applied when the total amount of corrections is below the maximum available refund volume (MARV). However, should the sum of the corrections in any given year exceed the MARV at the 66% refund rate, then the refund rate will be automatically reduced to a lower level generating total corrections equal to the MARV.

6.7. Financing rules

The financing of the UK rebate is based on the respective shares of the Member States in EU GNI. The UK itself does not participate in the financing of its own rebate. The Berlin European Council restricted the financing shares of the four net contributors (DE, AT, NL and SE) to 25% of their normal share.

Since the introduction of a generalised correction mechanism would limit and reduce the budgetary net balances of these countries, it does not seem warranted to maintain this distortion in the financing of the correction. The financing should therefore be simplified accordingly.

Technically the correction could be financed in three different ways:

- Member States who receive a correction do not participate in the financing. Under this option the entire burden of the financing would be concentrated on countries representing less than 50% of the total EU-27 GNI, many of them with relatively low levels of prosperity. Consequently, their net budgetary position would either deteriorate in an unacceptable manner, or the volume of the corrections would have to be severely restricted. Furthermore, distortions could result when a Member State with a net position just below the threshold would have to participate in the financing of the corrections and another Member State just above the threshold would be excluded from the financing.

- Member States would participate in the financing of all corrections except their own. This alternative would add considerable complexity to the proposal, since it would entail a separate financing round for each of the corrections.

- All Member States would participate in the financing of the total corrections. Their financing share in the global amount of the corrections would be determined by their share in GNI.

For sake of feasibility, transparency and simplicity, all Member States should participate in the financing of all corrections (third option).

Under these financing rules, the impact of adding the corrections to either the expenditure side or the revenue side of the EU budget would be identical. In both cases the corrections would be financed exclusively through an increase of the marginal resource GNI. However, adding the corrections to expenditure would artificially inflate the level of budgetary expenditure with a corresponding decrease of available margins under the global expenditure ceilings. That is why it seems preferable to maintain the GCM on the revenue side of the budget, as is currently the case for the existing UK correction mechanism.

6.8. General outline of the mechanism

On the basis of the above considerations, the calculation of the corrections stemming from a generalised mechanism would entail the following steps:

1. Calculating the aggregate amount of all expenditure allocated to each Member State. Allocated expenditure, as for the current correction mechanism, includes all categories of expenditure that are 'internal' to the EU (see section 6.2 above).

2. Calculating the percentage share of each Member State in total allocated expenditure.

3. Determining the percentage share of each Member State in own resources payments. Since traditional own resources are excluded from the categories of revenue to be taken into account, the combined share in VAT and GNI own resources would be used to calculate any Member States' share in EU budgetary revenue.

4. Deducting (2) from (3) above for each Member State, to obtain the percentage corresponding to the positive/negative balance.

5. For each Member State, multiplying the percentage resulting from (4) above by total allocated expenditure, to obtain the net budgetary contribution/benefit expressed in euros.

6. Multiplying the threshold corresponding to the pre-determined reasonable net contribution (RNC) by the GNI (in EUR) of each Member State and deduct from the result obtained under (5).

7. If the result obtained under (6) is greater than zero, multiplying the result under (6) by the refund rate to obtain the correction for each Member State and reduce the refund rate as necessary if the sum of all corrections exceeds the predetermined maximum refund volume (see section 6.5 above).

The total cost of the correction mechanism, i.e. the sum of all corrections would be limited by a maximum available refund volume (MARV), to be agreed ex-ante as a yearly pre-determined amount.

Net positions exceeding the RNC are eligible for a partial refund: the corrective percentage or the refund rate (RR), to be applied to the part of the Member State's net budgetary balance exceeding the threshold (RNC). This refund rate is a dependent variable for which an upper limit of 66% (currently used for the UK correction) applies. The actual refund rate will be derived from the available refund volume. Thus, the refund rate is automatically reduced when the application of the 66% maximum refund rate breached the MARV.

The partial refund is made ex-post as currently done for the UK correction.

The resulting formula for calculating the generalised correction mechanism is the following:

>REFERENCE A UN GRAPHIQUE>

=

>REFERENCE A UN GRAPHIQUE>

if

>REFERENCE A UN GRAPHIQUE>

> 0

where RR = 0.66 if

>REFERENCE A UN GRAPHIQUE>

>REFERENCE A UN GRAPHIQUE>

MARV and

RR =

>REFERENCE A UN GRAPHIQUE>

* 0.66 if

>REFERENCE A UN GRAPHIQUE>

> MARV

Where: // TP = Total VAT and GNI-based payments made by all Member States in respect of year t

// TPx = VAT- and GNI-based payments made by Member State x in respect of year t

// E = Total allocated expenditure in respect of year t

// Ex = Expenditure allocated to Member State x in respect of year t

//

>REFERENCE A UN GRAPHIQUE>

= Correction (at refund rate RR) to be granted to Member State x in respect of year t

// RNC = Reasonable net contribution (= threshold) expressed in % of GNI

// Yx = GNI of year t of Member State x

// RR = Refund rate (i.e. percentage of the excess contribution to be compensated)

// MARV = Maximum available refund volume

//

>REFERENCE A UN GRAPHIQUE>

= Total sum of corrections if RR is equal to 0.66

6.9. Comparing corrections levels

A threshold set at -0.35% of GNI will give rise to an estimated average volume of gross corrections of around EUR 7 billion for the period 2007-2012 (to be financed in 2008-2013).

The generalisation of the correction mechanism combined with the change in the financing rules means that the comparison with the level of the correction under the current system is somewhat blurred. As all Member States, including those benefiting from such a mechanism, will participate in the financing of all corrections, the overall net correction will always be lower than the gross correction, whereas under the current system there is no difference between net and gross correction (what the UK gets is what other Member States pay).

The resulting net balances of the large net contributors after the correction will be higher than the established threshold because of the combined effect of the partial refund and the participation in the financing of the correction system. Conversely, the burden for Member States having to pay fully their share of the financing cost and not benefiting from the correction is lower than at present even if the overall gross correction is larger. Thus, with a threshold set at 0.35% of GNI and a maximum predetermined correction volume set at EUR 7.5 billion, the combined burden for these Member State would be even lower (by about EUR 1 billion) than the average level paid during the 2001-2004 period and much lower (by about EUR 2.7 billion) when compared to the estimated cost of the current UK correction over the next financial framework.

6.10. A system based on the pooling of net balances

Extending from the original Fontainebleau principle, which inspired the current UK correction mechanism, a system could be envisaged that is not restricted to correcting the net balances of excessive net contributors but which would redistribute the net balances of all net contributors. The principal aim of such an approach would be to obtain a higher correlation between the net balances of the Member States concerned and their relative prosperity.

This could be achieved by first pooling the net balances of a specific group of Member States and, secondly, by sharing (redistributing) the total sum of the net balances among a (possibly different) group of Member States.

A number of alternative approaches could be envisaged. Choices to be made include:

* determining the group of Member States whose net balances would be pooled and determining which group of Member States would finance the sum of the negative net balances;

* deciding whether the financing should be proportional to GNI or if it should be progressive on the basis of GNI shares adjusted for relative prosperity.

In the alternative presented below net balances of net contributors closely reflect the PPS per capita ranking. However, it would entail considerable conceptual and technical complexity.

The system would pool together the net balances of the 12 Member States whose GNI per capita exceeds 100% of the EU-25 average. Over the period 2007-2013, 12 countries are expected to be in this category: UK, NL, DE, SE, AT, IT, FR, DK, BE, IE, LU and FI. The total sum of their net balances would be financed by the same 12 Member States. The financing is done on the basis of GNI shares adjusted for relative prosperity, which would be achieved by adjusting the GNI of each Member State in the group according to the relative level of prosperity expressed in PPS, and then recalculate the corresponding share of each Member State in the new total GNI.

This alternative would have major implications for Belgium, Luxembourg and Ireland, countries that for various reasons [35] are net beneficiaries from the EU budget, despite the fact that they have a relative prosperity above the EU average. To alleviate this effect, a second round of corrections is needed in which a further constraint would be added which limits the impact of the correction to a deterioration of 0.25% points of any Member State's net balance expressed as a percentage of GNI. This constraint reduces the contributions to the system from BE, IE and LU. These reductions are then financed by the other 9 Member States belonging to the group according to their share in GNI.

[35] In the case of Belgium and Luxembourg this is because they are allocated the bulk of the administrative expenditure and thus appear as net beneficiaries. Ireland receives a relatively high share in agricultural expenditure and also continues to benefit from some cohesion policies.

The table below illustrates the net balances for all Member States in comparison with their situation without any correction and under the current UK correction.

>EMPLACEMENT TABLE>

The basic effect of any correction mechanism based on the pooling of the net balances of all the net contributors, regardless of the parameters chosen, is to deteriorate - significantly - the net balances of those net contributors who are allocated a larger share of EU expenditure (more specifically agricultural and administrative expenditure) than the other net contributors. Mainly affected by this logic would be France, Denmark and Finland. Conversely, the current major net contributors (UK, NL, DE and SE) would all benefit accordingly. Moreover, under the parameters of this specific system, BE, IE and LU would see their net balance deteriorate by 0.25% points compared to the current system.

A correction mechanism based on pooling of the net balances has intuitive appeal, since all net contributors would end up with net contributions grouped within a narrower range (expressed as a percentage of GNI) and in correlation with their GNI per capita ranking. The fact that the poorer Member States would be excluded from the financing could be construed as an advantage from the point of view that relatively poorer Member States would not have to contribute to correcting the negative imbalances of relatively richer ones.

Besides the issue of the technical complexity and lower transparency, the system is however fraught with a number of conceptual disadvantages:

* A major threshold effect could theoretically occur: A net beneficiary with a relative prosperity just below the EU average would be excluded from the financing while a Member State with relative prosperity just above the average could participate up to the equivalent of 0.25% of its GNI.

* It would represent a departure from the Fontainebleau European Council conclusions, which evoke only the possibility to correct excessive net balances.

* The impact of such a correction mechanism, which would have a gross average volume of approximately EUR 40 billion annually, would be so important that it would fundamentally change the structure of the own resources system.

In theory this pooling system could be part of a mixed correction system, based on equal weighting of this pooling system and a GCM based on a threshold. Obviously such a solution would add further complexity.

7. The Commission's proposal

Any correction mechanism entails an additional complication to the financing of the budget. Therefore the basic proposal should be relatively simple and sufficiently transparent. In this respect, the current parameters of the UK correction calculation should be modified only where necessary and simplified whenever feasible.

The Commission therefore proposes to use the following parameters for a generalised correction mechanism:

- keeping the categories of revenue to be taken into account (VAT + GNI) unchanged. Any other existing complication [36] should be abolished;

[36] Such as the calculation of the 'UK advantage' and the TOR windfall gains.

- keeping the expenditure headings included in the allocated expenditure unchanged;

- the financing should be simplified and based only on GNI shares, whereby all Member States would participate in the financing of the global amount of the corrections on the basis of their relative prosperity;

- it is proposed to set the threshold level at -0.35% of GNI

- the refund rate will be the dependent variable with a maximum rate of 66%, to be reduced automatically when the agreed maximum refund volume is exceeded in a given year;

- the maximum available refund volume could be set at EUR 7.5 billion.

The amount of the gross corrections, their financing and the combined net effect are illustrated for all Member States in the annexed Tables X through XIV.

In absolute terms, the UK would be by far the largest beneficiary from the generalised correction mechanism, receiving on average an estimated net compensation in excess of EUR 2 billion per year, approximately twice as much as the net amount DE would receive.

The table below illustrates the estimated net budgetary balances for the period under consideration. (The results for all Member States are presented in the annexed Tables XIII, III and II respectively.)

>EMPLACEMENT TABLE>

With the proposed mechanism, the average net balances of the largest net contributors would be at comparable levels, with the UK, DE, NL and SE all between -0.51% and -0.45% of their respective GNI, while with the current system their estimated future net balances would range between -0.56% and -0.25% (and between -0.62% and -0.47% with no correction). Several Member States (FR, IT, CY and AT) would have net balances averaging between -0.40% and -0.30%, with AT somewhat higher than the other three. DK and FI would be the two remaining smaller net contributors with estimated average net balances of -0.19 and -0.25%, respectively.

Even with a threshold level of -0.35%, the net balances of the largest net contributors would end up at a higher percentage of their GNI because of the combined effect of the partial refund and the participation in the financing of the correction system. Nonetheless, the resulting net budgetary balances for net contributors would be more in line with the Fontainebleau principle when compared to the current system.

On the other hand, the estimated burden of financing the overall correction for all the other Member States would be lower than under the current system.

It would be appropriate that the introduction of the generalised system for correcting budgetary imbalances is accompanied by transitional measures for the UK to alleviate the financial impact of the changeover for this Member State. In order not to increase excessively the total cost of the corrections, the application of the generalised system should be progressively phased in for the other eligible Member States.

8. Conclusion

On the basis of its review of the own resource system, the Commission is of the view that the existing correction mechanism on an exclusive basis is no longer justified and proposes to introduce a generalised mechanism to correct excessive negative budgetary imbalances.

A generalization of the correction mechanism, evolving from the existing correction, would allow bringing the system closer to the original objective of avoiding excessive budgetary burdens. By introducing a sort of 'safety net' for large net contributors whose net contribution exceed a certain level, it may also facilitate a more constructive approach to ensure the budgetary means to meet the policy challenges of the enlarged Union.

The Commission proposes a generalization of the correction mechanism calculated on the basis of the net budgetary balance of each Member State in relation to the budget of the EU. The mechanism should be triggered if net contributions exceed a threshold, expressed as a percentage of each Member State's GNI, reflecting the minimum accepted level of unlimited financial solidarity between Member States. Net positions exceeding such a threshold will be eligible for a correction (partial refund), thus giving an insurance against excessive net contributions. Conversely, the total volume of corrections (refund volume) will be limited to a maximum amount, thus insuring those not benefiting from a correction against excessive costs of the mechanism. If the sum of all corrections exceeds the total predetermined volume, the refund rate is reduced accordingly.

The proposed generalized correction mechanism will decrease the negative net balances, reduce the spread among net contributors, and, on the other hand, lessen the financing burden of those who do not benefit from the mechanism.

PART III - THE OWN RESOURCES STRUCTURE

1. Finding the optimal own resources structure

The pros and cons of alternative financing systems are analysed in light of the preceding assessment of the current financing system. Three basic alternatives for the financing of the EU budget are discussed below:

(1) Maintaining the present financing system unchanged;

(2) A purely GNI-based financing system;

(3) A financing system based on fiscal own resources.

All alternatives examined in the sections below assume that TOR would continue to be part of the own resources, as the collection of import duties [37] constitutes an instrument of the Union's trade policy whose yield 'naturally' accrues to the Union. The assessment will focus on own resources other than the TOR.

[37] TOR also include sugar levies. These are, however, not levied on imports but on sugar producers in the EU for the purpose of financing sugar export refunds. These levies amount to less than 1 % of total own resources.

1.1. Maintaining the present financing system unchanged

It could be argued that despite its weaknesses in terms of complexity, opacity, limited autonomy from national treasuries and the European Parliament's limited political accountability for its expenditure decisions, the present financing system has ensured a smooth financing of the EU budget.

However, in its present form the financing system is unable to give visibility to the budgetary consequences of the Union's policies to the general public. The UK correction mechanism has become increasingly and exceedingly complex over time, adding to the opacity of the system. Furthermore, due to the tendency of the UK correction to increase with the present and future enlargements the system will become less equitable over time. The possibilities for marginal adjustments have reached their limit. The system unmistakably contains an increasing incentive for Member States to adopt a narrow 'juste retour' stance, which influences negatively the decision-making process in the Council. Member States, in particular net contributors, tend to judge the merit of new Community initiatives exclusively in terms of their financial consequences with little regard to the substance of policies, leading to the risk of obscuring the added value of EU policies.

The present system resembles very much a GNI-based system without offering the advantages in terms of cost-effectiveness and simplicity that would derive from a system based exclusively on GNI (see section below). The small part of the income corresponding to the current 'statistical' VAT resource will remain. Member States will be confronted with the complex and time-consuming task of reconstructing the VAT base at a time when the share of VAT in total financing only represents around 15 %. Moreover, the VAT bases of around half of the 25 Member States will most likely exceed the 'capping' limit of 50 % of their GNI. This turns the VAT contribution into a de facto GNI contribution, as any increase in VAT receipts in a Member State whose VAT base is 'capped' will have no effect on the EU VAT resource.

1.2. A purely GNI-based financing system

This scenario would mean pursuing the trend observed since the late eighties to its logical extreme, abolish the current statistical VAT resource and finance the budget entirely by GNI-based contributions from Member States (apart from the small share of current TOR that would stay).

The suppression of the statistical VAT resource, with its specific rules and corrections, would simplify the present system. Sufficiency and stability of revenue within the agreed own resources ceiling would be guaranteed, with low operating costs.

Under this scenario the Union would almost entirely depend on 'contributions' from Member States. The advantage of such a system is that it would be simple and easy to understand. The 'contributions' would correspond closely to Member States' relative prosperity. However, financing the budget by contributions of the Member States is adequate for an international organization such as the United Nations, but it does not reflect the status of the European Union.

It would imply an idea of the Union in which citizens would be represented purely indirectly by their Member States. The status of the EU as a Union of Member States and the citizens, which is currently reflected in the Treaty, would be abandoned on the financing side of the budget. This, in turn, would imply ditching the concept of 'own' resources, which has been a cornerstone of EU finances since the first own resources decision in 1970.

As a consequence, the debate on fair burden-sharing among Member States and 'juste retour' would be brought to the forefront of the European debate even more than at present. There would be virtually no visibility of the financing for EU citizens and it would risk precluding any future re-opening of the debate on fiscal revenue replacing national contributions.

1.3. A financing system based on fiscal own resources

Under this scenario all or a large part of the EU budget would be financed from tax-based resources paid by citizens and economic operators. All or a large part of the GNI resource and the entire current 'statistical' VAT resource paid by Member States would be replaced by one or several fiscal resources.

This approach would allow for increased financial autonomy of the EU budget, by assigning (part of) taxes to the EU level, preferably those characterised by a high level of regional arbitrariness [38].

[38] Regional arbitrariness refers to a situation where it is difficult to determine the exact share of a tax base to be allocated to individual Member States or where there is a high (potential) mismatch between the country collecting the tax and the country of residence of the economic agents bearing the burden of the tax.

Financing the budget by tax-based resources would also allow for a more direct financial relationship between the EU budget and EU citizens. This would imply a shift towards individuals and economic operators as contributors, and the corresponding reduction of Member States' contributions, in the financing of the EU budget. Such an approach would entail increased visibility of the financing of the EU budget to the citizens. It would also contribute to shifting the political discussion away from the narrow, and often ill-defined, focus on national interest contributions towards the merits of EU policy and the general European interest.

As citizens/taxpayers tend to question the use and the amount of taxes they pay, a direct link between the financing of the budget and the citizen would thus induce increased political accountability of the budget authority for its spending decisions. This in turn could have positive consequences in terms of efficiency, budgetary discipline, and transparency. Within the limits delegated by the own resources decision, income responsibility might be given to both arms of the budget authority, thus increasing accountability of the European Parliament towards EU taxpayers.

A financing system based exclusively on fiscal own resources would mean that a fiscal resource would take over the current role of GNI as 'residual' balancing resource [39], involving a variable tax rate according to budgetary requirements. Thus, the adoption of the annual budget, as well as any increase in budgeted expenditure during budget execution, would entail a variation in the call rate on the residual tax resource through changes in the tax rate or require a higher flexibility on the expenditure side than the current financial perspective framework allows. However, frequent amendments to the tax rate would in practice be very unwelcome by taxpayers (legal uncertainty, technically and administratively cumbersome procedures) and could conflict with national tax-setting. This speaks in favour of maintaining a residual balancing GNI-based resource. The GNI based resource could continue to play such a buffer role also in this scenario, although with a significantly reduced weight in total financing. Alternatively, the possibility of financing budget deficits through borrowing from the financial markets would be necessary (combined with the possibility of running surpluses too). However, that would require departing from the principle that 'revenue and expenditure shown in the budget shall be in balance' as re-affirmed in the constitution (article I-52) agreed by the inter-governmental conference in June 2004. This principle should not be put into question.

[39] That is, offsetting the difference between total expenditure and other revenue to ensure a balanced budget.

The introduction of fiscal own resources relies on a sufficient harmonisation of tax bases at EU level (indispensable for equity reasons). In some areas significant progress has already been made (e.g. VAT and mineral oils) whereas in other areas steps towards harmonisation have been much more limited. Further progress in this field depends upon the political will to achieve such harmonisation. In recognition of the political and institutional constraints, ambitions for what can be achieved for the next financial framework have to be limited.

1.4. Conclusion

The present financing system works and has ensured a smooth financing of EU expenditure. However, if left unchanged it will perpetuate its shortcomings in terms of complexity, opacity, lack of visibility to EU citizens, deficient equity and excessive dependence on transfers from national treasuries, with the result of exacerbating Member States' narrow focus on 'juste retour'. This may undermine support for policies pursuing the general European interest and the integration process. The introduction of a generalised correction mechanism (see Part II above) would address the lack of equity and the unsustainability of the existing correction mechanism which needs urgent remedy. However, a generalised correction mechanism would not bring a direct link to citizens/taxpayers nor reduce the EU budget's growing reliance on transfers from national treasuries.

Moving to a system based only on the GNI resource would mean that most of the drawbacks and deficiencies linked to the current system would be further reinforced. It would depart from the idea of a Union of Member States and the citizens of Europe and would imply abandoning the very idea of fiscal 'own' resources.

A system based to a large degree on tax-based own resources has the potential to increase the financial autonomy of the EU budget and to create a more direct link to EU citizens. It could also contribute to shifting the attention towards the merit of the EU policy and the general European interest. Some of the candidate taxes may also contribute to a more efficient allocation of resources within the internal market. However, at this stage of the EU integration process, a fully tax-based financing of the EU budget does not appear realistic. A progressive approach, consisting in maintaining a limited GNI resource while increasing the share of tax-based resources, appears preferable.

2. Towards a financing system based on contributions by Member States and citizens

2.1. Increasing the share of tax-based own resources

The previous analysis shows there is a need to reform the current financing system by giving a larger role to tax-based own resources. A reformed own resources system could combine additional tax-based own resources and the existing GNI-based resource as the residual resource.

At the same time, the current 'statistical' VAT-based resource should be abolished. It is in practice a resource levied on Member States and not on citizens that imposes an additional workload on national administrations and complicates the financing system, without providing any advantages compared to the GNI-based resource.

As the financing system is to be based on several own resources, there is no need for each individual resource to meet all assessment criteria. Indeed, there is no such thing as a perfect own resource able to meet all criteria simultaneously. Each resource has advantages and disadvantages and performs well according to some criteria and less well according to others. In short, it is the overall system, rather than each individual own resource, that should be assessed against the criteria mentioned in Part I above.

In particular, the reform of the financing system should be aimed at overcoming the drawbacks of the current system, i.e. the absence of a direct link to EU citizens, dependence on transfers from national treasuries and unjustified complexity, while contributing as far as possible to an efficient allocation of resources. Other criteria become less relevant for the new resources, as long as the overall system allows them to be met to a reasonable degree.

For reasons of equity between citizens in different Member States, the introduction of new fiscal resources requires a sufficient prior harmonisation of the tax base. The choice of the most appropriate tax-based resource crucially depends on the actual degree of tax base harmonisation.

The existence of regional arbitrariness [40] or the presence of cross-border externalities are strong arguments for harmonising tax bases and rates, which could also justify assigning the corresponding tax revenue to the higher EU level.

[40] See footnote 11.

Increasing the share of tax-based own resources in the financing of the EU budget does not require any new taxes. It does, however, imply a decision on sharing either revenue or tax rates between the national or the EU level. The EU share could be levied as part of the national rate paid by taxpayers. The total EU budget is in any case limited by the own resources ceiling to a maximum of 1.24 % of EU-GNI, and would not increase, as revenue from the tax-based resource would be offset by a corresponding decrease of the current GNI-based resource.

The following sections illustrate three main options in order to improve the functioning of the EU financing system. All three options retain the current GNI-based resource as a residual balancing resource as well as the traditional own resources, but assume that a significant part of the budget is financed by a new fiscal resource. It is assumed that the new resource would also replace the current statistical VAT. The exact percentage of the budget to be financed by fiscal resources is a matter of political choice and acceptability but as a working hypothesis it has been assumed that the new fiscal resource would cover around half the EU budget.

A common assumption for the three options is that the total tax burden should not increase as a result of a reform of the EU financing, since any increase in EU fiscal resources is offset by a corresponding decrease in the GNI-based resource called in from Member States. This gives leeway to Member States for corresponding tax decreases.

The first two options concern consumption-related own resources, based on energy consumption and on transactions subject to value added tax (VAT), respectively. Both of these potential resources, and particularly VAT, would allow for a direct and visible link to EU citizens. It could, furthermore, be seen as 'natural' to levy certain energy taxes at the EU level because of the mismatch between the geographical pattern of tax collection and tax burden, which makes any national reapportioning arbitrary. The degree of harmonisation at the EU level has made good progress in both of these areas.

The third option concerns revenue from an EU levy on company income. A single EU company tax base and system for all companies in replacement of the existing national systems would contribute to the proper functioning of the internal market by contributing to a level playing field for foreign direct investments and by further enhancing the potential for cross-border activities. Such a tax-based resource would be closely related to the need for a general harmonisation or at least approximation of company taxation in the EU in order to eliminate fiscal obstacles to the proper functioning of the internal market.

It is apparent that, for any of the three options, the pace of progress towards harmonisation and for the technical preparatory work related to the introduction of a new fiscal resource will crucially depend on the underlying political will.

It is also worth mentioning that the introduction of new fiscal resources, especially if they involve tax bases involving regional arbitrariness, should entail a reassessment of the justifications for any mechanism of correcting budgetary imbalances (see Part II of this report).

An overview of other potential candidates for own resources that have not been retained in any of the three options can be found in the 1998 Commission report on the operation of the own resources system (annex 2) [41].

[41] Agenda 2000, Financing the European Union, Bulletin of the European Union, Supplement 2/98, also available at http://europa.eu.int/comm/budget/agenda2000/reports_en.htm. An update of such an assessment is to be found in 'Tax-based EU own resources: An assessment', Working Paper no. 1/2004, Taxation Papers, Directorate General Taxation & Customs Union, European Commission.

2.2. Option 1 - An own resources system with fiscal resources related to

energy consumption

Under the new directive on energy taxation [42], which entered into force on 1 January 2004, most energy products are subject to Community taxation. The directive widens the scope of the Community-wide minimum rates of taxation, previously limited to mineral oils, to all energy products, such as natural gas, electricity and coal and provides for harmonisation of tax bases.

[42] Directive 2003/96/EC of 27.10.2003 of the Council restructuring the Community framework for the taxation of energy products and electricity (OJ L 283 of 31.10.2003).

Although the adopted directive on energy taxation was not prepared for the purpose of introducing new own resources, it nevertheless creates suitable conditions by harmonising tax bases and establishing minimum rates.

A fiscal resource on energy products would, however, not need to be based on all products covered by the directive. It would be sufficient to set an EU levy on the tax base related to motor fuel used for road transport. This would be relatively simple from an administrative point of view as the tax base is already harmonised at the EU level and since most possibilities of tax differentiation allowed by the directive apply to other products, such as heating fuel and electricity and to some specific uses of motor fuel. A possible further development might include a levy on aviation fuel or related emissions.

2.2.1. Outline of an energy-based fiscal resource

The tax base for taxation of motor fuel used for road transport is already harmonised at EU level. Motor fuel used for road transport mainly includes leaded and unleaded petrol, diesel, LPG and natural gas used for transport. Energy products used as motor fuel for certain industrial and commercial purposes would thus not be subject to the EU levy.

The main issue for discussion would be the level of taxation in the context of defining an own resource as well as the concrete operation of the system, including the practical arrangements for the possibility of displaying the EU levy on receipts and invoices. The EU rates used for the own resources revenue collection could be set at levels equivalent to the minimum rates defined in the energy taxation directive or at different levels. EU rates below half of the minimum rates would be enough to finance half of the current EU budget.

In this context it could be mentioned that for at least one subcategory of motor fuel for road transport there is a strong case for complete harmonisation also of the tax rates at EU level. Diesel used for professional transport is a tax base with a significant degree of mobility mainly due to the possibility of hauliers and coach operators engaged in international activities to take advantage of the very significant differences in national excise duties on diesel by filling up in Member States where prices are lower [43].

[43] 'Big trucks have tanks of huge capacity, which allow them to cover between 1 500 and 3 000 kilometres on a single tank. That means in reality that hauliers involved in international activities conduct a kind of fiscal planning: they take advantage of the very significant differences in national excise duties on diesel by filling up in Member States with the lowest taxes.' (Commission proposal for a Council directive to introduce special tax arrangements for diesel fuel used for commercial purposes, COM(2002) 410 final of 24 July 2002). See also a study by OECD, (OCDE/GD(97)69: CO2 emissions from road vehicles).

A single EU-wide rate on diesel used for professional transport, as already proposed in the Commission proposal for a Council directive to introduce special tax arrangements for diesel fuel used for commercial purposes [44], would avoid distortions in the internal market.

[44] COM(2002) 410 final of 24 July 2002.

It may be interesting to compare the distribution of potential revenue from a levy on motor fuel for road transport across Member States. The table below compares the shares of Member States in the consumption of motor fuel used for road transport with their shares in EU GNI.

>REFERENCE A UN GRAPHIQUE>

Seen from a Member State perspective, differences between the share in total EU consumption of motor fuel for road transport and the share in GNI are very substantial (see columns 7 and 8). This is the case, in particular, for several of the new Member States, such as Estonia, Latvia, the Slovak Republic, Lithuania or the Czech Republic. The high share of Luxembourg is partly related to consumption by non-residents taking advantage of the lower tax rates on motor fuel in that country. However, a uniform EU rate, integrated in the national tax rate and applied in all Member States, would be levied on consumers and not on Member States. Seen from a consumer perspective, the impact would be the same on the comparable consumer across the Union.

An EU levy on aviation fuel or related emissions might be a useful complement to a levy on motor fuel for road transport. The European air transport system is highly integrated and aviation emissions transcend national borders. There would therefore be a certain logic in attributing revenue stemming directly from the taxation of such emissions or from aviation fuel taxes to the EU level. It would also be a way of internalising the external socio-economic costs of climate change and other environmental effects caused by aviation into the price of air travel. Contrary to other motor fuels, aviation fuel (kerosene) is currently exempted from taxation for cross-border flights, which gives a competitive advantage to this mode of transport.

The coverage of an emission charge or a fuel tax may range from national flights only to all flights for all carriers to all destinations worldwide (from the EU). Whereas one may assume that all intra-EU flights would be submitted to a system of emission charges or fuel taxes, the inclusion in the scheme of flights departing from or arriving in the EU would require further steps and decisions.

A specific difficulty related to a possible aviation levy is that, unlike other existing taxes, it would mean taxing a base that is currently not taxed, and might therefore be perceived as an additional tax burden rather than reflecting a different sharing between national and EU level of existing tax revenue.

Policy discussion on aviation taxation is currently evolving. For the time being, there are ongoing discussions on the idea of emissions-trading related to aviation, on the taxation of kerosene for aviation in the context of open methods of coordination or enhanced cooperation, and with regard to the own resources. Particular attention should be paid in the future to the links between these issues.

Revenue estimates

From an administrative point of view, setting an EU levy on motor fuel used for road transport would be feasible. The main decision would bear on the precise rates to apply. Potential revenue would be abundant. EU rates below half of the minimum rates in the energy taxation directive would be enough to finance half of the EU budget (see also table 1 above). In practice, the EU rate could vary according to product and use. Finally, since the levy would be based on quantities sold and not ad valorem, and due to the low price elasticity of demand, revenue would be relatively insensitive to the price of energy on international markets.

Estimates of revenue from aviation emissions charges are surrounded by a large degree of uncertainty, but ultimately depend on the charges deemed necessary to internalise the external costs of climate change and other negative environmental effects of aviation. Other political and economic considerations may obviously also come into play. Based on data for EU-15 [45], an aviation emissions charge limited to internalise the negative external climate costs of CO2 and NOx emissions could bring up to around EUR9 billion a year. Should other emissions be taken into an emissions charges scheme, e.g. water responsible for the condensation trail, amounts could be substantially higher, maybe up to EUR20-25 billion.

[45] Wit and Dings, 'Economic incentives to mitigate economic greenhouse gas emissions from air transport in Europe', July 2002.

Introducing an EU levy of EUR330/1000 litre (the minimum level for non-exempted kerosene in 2010 according to the energy taxation directive) would bring up to around EUR17 billion, according to preliminary estimates based on actual consumption in 2001 in the current 25 Member States.

However, two important caveats have to be made. First, proceeds from aviation taxation (charge or fuel tax) would be substantially lower if the tax covers only intra-EU routes. Second, imposing aviation fuel taxation may face difficulties due to bilateral Air Service Agreements (ASA's) which exist between individual Member States and other countries. The issue is being addressed in the ongoing reform of the Community's external aviation relations but the state-of-play would need to be taken into account in designing an aviation taxation scheme.

Time-table for introducing the new resource

Since the possible harmonised tax base would be directly drawn from the energy taxation directive and since there exist minimum rates that could constitute the basis for the EU own resource rates, only relatively limited technical work would be required in order to develop a resource based on motor fuel for road transport.

The following main steps would be required:

- Decision as to the base, i.e. define the energy products and use covered by the directive that should come into consideration for EU taxation,

- Decision as to the EU rates,

- Rules regarding the information to the taxpayers (element of the tax paid accruing to the EU budget vs. the national budget),

- Legal and administrative rules on the collection and allocation of the tax proceeds.

In a first analysis, it could be estimated that between 2 and 4 years would be required to design this fiscal resource. Since a transition phase would be required to allow economic operators to adapt, a reasonable estimate for the launch of the EU levy may be 3 to 6 years.

As for a resource based on aviation fuel or emissions, between 2 and 3 years could be necessary for the technical preparation. However, the existence of the infrastructure of the European Organisation for the Safety of Air Navigation (EUROCONTROL) and the nature of taxpayers involved could facilitate operational implementation [46]. Overall, about 4-5 years may seem a reasonable estimate for the lead time required before a system of aviation fuel taxes or emissions charges could come into effect.

[46] An agreement would in that case have to be reached with EUROCONTROL to use their infrastructure as a vehicle for collecting the tax.

2.3. Option 2 - An own resources system with a fiscal VAT resource

This option focuses on introducing a fiscal resource that can serve as a direct link between the citizens and the EU budget in order to increase visibility and financial autonomy.

The main objective of a fiscal VAT-based own resource would be to enhance the link with citizens, rather than with Member States or economic operators (see above), as EU taxpayers. The application of an EU rate to national VAT bases would give visibility to the financing of the EU budget, increase awareness of the costs of the Union and contribute to a better identification of citizens with Europe. For this, the VAT resource ought to be levied directly on the national VAT base rather than on the theoretical VAT base which would exist under a fully harmonised and uniformly applied system, as done at present.

As explained in part I of this report, the current VAT-based resource is not a truly fiscal resource. It has only an indirect link with national VAT, and thus with the taxpayer/citizen. The VAT paid by citizens constitutes only the base on which statistical methods and macro-economic data are applied to arrive at the VAT-based resource. It therefore means sharing the tax revenue derived from a statistical tax base. Despite being formally a tax on citizens' consumption, the current VAT-based resource has the character of a national contribution since in practice it is levied on Member States and not on citizens.

2.3.1. Outline of a genuinely fiscal resource based on VAT

A genuinely fiscal VAT resource would be implemented through an EU rate as part of the national VAT rate paid by taxpayers. It would imply a specific percentage rate of VAT that would be levied for the benefit of the EU. The rate would be incorporated in, and levied together with, the national rate and thus on the same taxable base. Citizens would not have to support an additional tax burden as the Community rate would be offset by an equivalent decrease of the national VAT rate. For example, if the national VAT rate is 21 %, and assuming the introduction of an EU rate of 1 %, the national rate would come to 20 %. The total VAT rate levied would still be 21 %.

For visibility purposes, the Community VAT and national VAT should appear as separate taxes on the invoice or receipt that a taxable person provides to his customer. There would be no need to make a similar distinction in the tax returns.

Since different VAT rates are applied across the EU, Member States would transfer a different share of their VAT revenue, but the revenue transferred would correspond to the same percentage of each national VAT base.

Payments would be made to the EU budget as and when the tax was collected nationally. The corresponding amounts would not necessarily pass through the national budgets, which would merely record the balance of VAT collected after deduction of the direct contribution to the EU budget. This would enhance its nature as an EU 'own' resource and establish a direct link between the Union and its citizens.

Such a system may require a modification of Community VAT legislation as regards the elements that tax returns and invoices should contain, the tax periods and the deadlines for submission of tax returns and resulting payments.

Other issues that would have to be taken into account when designing a genuinely fiscal VAT-based resource include the possible uneven burden sharing. This is related to incomplete harmonisation (in particular zero-rated supplies [47]) and the tendency for the VAT base to be higher relative to national income in less prosperous Member States (the so-called regressivity of VAT), which since 1988 has pushed Member States to reduces its share in own resources.

[47] Zero-rated transactions refer to transactions that are not subject to output VAT but entitles to deduction of input VAT. Where national administrations apply a zero rate, it is difficult to apply any EU rate to be compensated for by an equivalent reduction in the national rate. Indeed, the rate in these cases would have to remain zero and, therefore, corresponding revenues would be nil. Not applying an EU surcharge at all to these goods would lead to important national differences, while applying only the EU rate to these zero-rated goods would lead to considerable administrative and political problems. Alternatively, a higher EU VAT rate could be levied generally in the concerned countries, or a financial compensation calculated, in order to offset the effect of zero-rated transactions.

The table below compares Member States' shares in the sum of national (unharmonised) VAT bases with their shares in total EU GNI. The relative importance of zero-rated transactions is also indicated (see column 4).

>REFERENCE A UN GRAPHIQUE>

Seen from a Member State perspective there is, in several cases, a considerable difference between the share in a fiscal resource based on VAT and a levy based on GNI (see columns 7 and 8). In general, countries that benefit from the current capping rule (i.e. countries for which the current harmonised VAT base exceeds 50 % of their GNI) do, not surprisingly, have a much higher share in a fiscal VAT based on the national VAT bases compared to their share in EU GNI. Inversely, the presence of zero-rated transactions tends to decrease the VAT base, as can be seen in the case of the United Kingdom. On the other hand, for countries such as Cyprus or Malta the effect of the zero-rates is more than compensated by the high share of (taxable) consumption in national income. However, a uniform EU rate applied in all Member States would be levied on consumers and not on Member States. Seen from a consumer perspective, the impact would be the same on the comparable consumer across the Union, except where zero rates are applied. Several possibilities of off-setting the impact of zero-rates can be envisaged.

The following table compares Member States' shares in the sum of national (unharmonised) VAT bases with their shares in the current harmonised and capped EU VAT base.

>REFERENCE A UN GRAPHIQUE>

Seen from a Member State perspective there is, in several cases, a considerable difference between the share in a fiscal resource based on VAT and the share in the current harmonised and capped 'statistical' VAT resource (see columns 7 and 8). As in the above comparison with GNI, countries that benefit from the current capping rule (i.e. countries for which the current harmonised VAT base exceeds 50 % of their GNI) generally have a much higher share in a fiscal VAT compared to their share in the harmonised 'statistical' VAT base. Conversely, the presence of zero-rated transactions tends to decrease the fiscal VAT base, which is clearly seen in the case of the United Kingdom. However, for some countries, the variations go in different directions depending upon whether the fiscal VAT is compared with GNI or with the current harmonised 'statistical' VAT resource [48].

[48] There may be several reasons for this. Apart from the impact of the capping rule and the zero-rated transactions the comparison is also influenced by the corrections to gross VAT receipts and positive or negative compensations added to the national VAT base in order to establish the harmonised base. Furthermore, the relation between the VAT base and GNI may vary considerably also among the non-capped countries.

Revenue estimates

VAT is a buoyant source of revenue representing on average around 7 % of GDP in the current 25 Member States during the period 1995-2002 [49]. Assigning to the EU level a rate of 1 % of the existing national VAT bases of all Member States would be enough to finance roughly half of the EU budget.

[49] Structures of the taxation systems in the European Union, Data 1995-2002, European Commission 2004.

Time-table for introducing the new resource

Harmonisation in the area of VAT is quite advanced. However, there are several issues outstanding that would need to be discussed in parallel. The first set of issues bear on the adjustment of the existing system. In particular, decisions would need to be taken on the following issues:

- the treatment of zero-rated goods,

- the treatment and the possible harmonisation of exemptions and restrictions to the right to deduct,

- (possibly) the improvement of collection mechanisms.

Secondly, concerning the EU VAT as such, decisions would need to be taken on the following topics:

- the EU rate and which goods to submit to this rate,

- the requirements for the economic operators (information on EU and Member States' VAT on invoices, etc.),

- the collection of VAT revenues and their transfer to the EU budget.

In principle, all these issues could be discussed and negotiated in two or three years. The necessary modifications of national rules and their implementation, as well as a transition phase to allow a smooth adaptation of economic operators, could require an additional two to three years.

In total, a reasonable estimate might be that it would require up to 6 years to actually put in place a fiscal VAT assigned to the EU level.

2.4. Option 3 - An own resources system with a fiscal resource based on

corporate income

As for the previous two options, also an EU fiscal resource based on corporate income would require the definition of a common (consolidated) tax base, in this case for companies. Such a harmonisation would contribute to the proper functioning of the internal market and a more efficient allocation of economic resources due to the cross-border externalities observed in the area of company taxation. Today, the existence of 25 separate national tax systems and the multiplicity of tax laws, conventions and practices represent in themselves a barrier to cross-border economic activity. They impose substantial compliance costs on companies operating across borders in the EU and lead to numerous loopholes in the tax system.

2.4.1. Outline of a possible fiscal resource based on corporate income

The main obstacle to be overcome for using corporate income tax as an EU fiscal resource is the existing high degree of diversity across Member States as regards the taxable base (rules on depreciation, deduction of losses, inventories, dividends etc.). A common EU rate on Member States' existing bases would lead to an inequitable distribution of the tax payments between companies as well as countries.

Current work on a comprehensive reform of company taxation is focusing on the concept of a common (consolidated) tax base, in all likelihood for a sub-group of interested Member States, to remove tax obstacles to the internal market. This work does not envisage any action on tax rates, nor is it conceived as a method for raising revenues for the EU budget. Optional schemes, which is currently discussed as one possibility, whereby companies can choose between belonging to the EU scheme or belonging to the national system would not be suitable in the context of own resources, since that could lead to 'vertical' tax competition between the EU and the Member States.

The option of a corporate income tax as an EU resource would imply setting a minimum tax rate to the harmonized tax base.

Due to limited data available it is difficult to properly evaluate the impact across Member States of a possible EU levy on a harmonised tax base. However, it may be useful to present Member States' current revenue from taxes on company income in order to show the impact by Member State of the potential loss of national revenue if taxation of this base was transferred to EU level.

The table below shows collected taxes on corporate income in the current Member States in % of GNI over the period 1995-2002.

>REFERENCE A UN GRAPHIQUE>

Although the average for EU-25 ranges between 2.08 % and 2.83 % over the period, there are much larger differences between Member States (see table above). During the period under consideration, revenue from taxes on corporate income has been comparatively low for instance in Slovenia (1.00% of GNI on average), Lithuania (1.02%) and Germany (1.15 %), and comparatively high for example in Luxembourg (7.84%), Cyprus (4.57%), the Czech Republic (4.08 %), the UK (3.27 %) and Italy (3.12 %), whereas France (2.44 %), Spain (2.77 %) and Poland (2.63 %) are close to the EU average. Variations over the period are also significant.

Revenue estimates

Corporate income tax represented on average 2.6 % of total EU-25 GNI during the period 1995-2002, which by far exceeds the financing needs of the EU budget. It should be noted that revenues from corporate taxation are highly variable over time as a result of the business cycle. In an EU financing system composed of a tax on corporate income as well as of GNI-based contributions (and traditional own resources) such variations would, however, be compensated by the residual balancing GNI-based resource.

Time-table for introducing the new resource

It is particularly difficult to draw a clear time frame for an EU fiscal resource based on corporate income. Indeed, the degree of harmonisation already achieved in the area of energy taxation and VAT at EU level is considerably higher than in the corporate income tax area.

The implementation of an EU own resource based on corporate income would first require an in-depth preparation on the following issues:

- the definition of a possible EU consolidated tax base,

- the implications for the current network of bilateral tax treaties linking EU Member States with other countries,

- the implications for national rules regarding personal income taxation.

In a second stage, a number of further practical issues would need to be examined, in particular:

- the rate of the tax and the allocation of the base or the revenue between the Member States,

- the responsibilities regarding the collection of the tax and related legal and administrative issues.

Part of the latter preparations could be done in parallel to the issues mentioned for the first stage above. It would then be necessary to count on an adjustment period of one to two years to allow national legislators to modify existing rules and effectively implement the tax.

Provided that the necessary political consensus exists it might be reasonable to estimate an effective preparatory period of between 5 to 10 years before a fiscal resource based on corporate income could be implemented. A road map for implementation would first require a political agreement on the principle of achieving harmonization of the tax base. A target date could, for example, be set by the European Council as an element of the decision on the next financial perspective.

2.5. Conclusion

The introduction of a new tax-based own resource replacing the current statistical VAT-based resource and financing a significant part of the EU budget would make it possible to overcome the main drawbacks of the current system, i.e. the absence of a direct link to EU citizens, overwhelming dependence on transfers from national treasuries and unjustified complexity. It could also contribute to a better allocation of economic resources in the EU. The existing traditional own resources should remain as well as the GNI-based resource as the residual resource. Even if accounting for a lower share of total own resources than under the current system, the GNI-based resource should continue to play an important role in order for the overall system to be able to reasonably fulfil all the relevant criteria.

The Commission proposes three main candidates as possible future fiscal own resources: a resource based on 1. energy consumption; 2. national VAT bases; and 3. corporate income.

A resource based on energy consumption and conceived as an EU levy on motor fuel for road transport would be a sufficient and stable financing source for the EU budget and would create a direct link to the citizens. The tax base is already harmonised at the EU level. An EU levy on aviation fuel or the related emissions could also be envisaged as a possible future development to end the current tax exemption for jet fuel and set a price on the environmental costs of aviation.

Tax base harmonisation in the field of VAT is quite advanced and it is a buoyant and stable source of revenue. A fiscal VAT resource would make the financing of the EU highly visible to EU citizens. It would also be evolutionary, since it would entail a reform of existing provisions rather than the introduction of a completely new resource. From an administrative point of view, its introduction would not present any insurmountable difficulties.

Due to the link to a common EU policy and the presence of cross-border externalities, revenue from a harmonised company tax base would also be a suitable financing source for the EU budget. Potential revenue is abundant, but varies with the business cycle. However, any shortfall would automatically be compensated by the residual balancing GNI-based resource. Unanimous agreement on the introduction of an EU fiscal resource levied on corporate income would also give additional momentum to harmonisation of the tax base.

The European Union is a Union of Member States and citizens. Assuming a financing share of half the EU budget, and due to the direct link with citizens/economic operators, any of the three candidates for fiscal own resources examined above would transpose this concept into the area of financing the EU budget. Strengthening the direct link between citizens and the EU budget would help focusing expenditure debates on substance rather than on purely "national" budgetary "net positions". It would also entail higher visibility and thus increased political accountability of the budgetary authority for spending decisions.

Any new assignment of a resource to the EU budget has not only to be decided unanimously by Council, but must also be ratified by all Member States' parliaments.

From an administrative perspective, the implementation of energy- and VAT-based resources would be feasible over the medium-term, whereas a fiscal resource based on corporate income is to be seen as a much longer-term option.

What is needed now is a political orientation to prepare the conditions for reforming the structure of the existing own resources.

In this regard, the Commission invites the Council to discuss the options proposed in this report and take note of the Commission's intention to prepare a roadmap in view of replacing, on the basis of a Commission proposal, the current VAT resource by a genuinely tax-based own resource by 2014.

Part IV - General conclusions

The European Union is a Union of Member States and citizens. Assuming a financing share of half the EU budget, and due to the direct link with citizens/economic operators, any of the three candidates for fiscal own resources examined above would transpose this concept into the area of financing the EU budget. Strengthening the direct link between citizens and the EU budget would help focusing expenditure debates on substance rather than on purely "national" budgetary "net positions". It would also entail higher visibility and thus increased political accountability of the budgetary authority for spending decisions.

From an administrative perspective, the implementation of energy- and VAT-based resources would be feasible over the medium-term, whereas a fiscal resource based on corporate income is to be seen as a much longer-term option.

What is needed now is a political orientation to prepare the conditions for reforming the structure of the existing own resources. In this regard, the Commission invites the Council to

- discuss the options proposed in this report,

- take note of the Commission's intention to prepare a roadmap in view of replacing, on the basis of a Commission proposal, the current VAT resource by a genuinely tax-based own resource by 2014.

In order to provide a short-term solution to the issue of excessive budgetary imbalances, the Commission is of the view that

- the existing correction mechanism on an exclusive basis is no longer justified and proposes to introduce instead a generalised mechanism to correct excessive negative budgetary imbalances.

In this context it is important to stress that the introduction of a new fiscal resource, as discussed above, would probably require a review of the generalised correction mechanism.

A generalization of the correction mechanism, evolving from the existing correction, would allow bringing the system closer to the original objective of avoiding excessive budgetary burdens. By introducing a sort of 'safety net' for large net contributors whose net contribution exceed a certain level, it may also facilitate a more constructive approach to ensure the budgetary means to meet the policy challenges of the enlarged Union.

The Commission proposes a generalization of the correction mechanism calculated on the basis of the net budgetary balance of each Member State in relation to the budget of the EU. The mechanism should be triggered if net contributions exceed a threshold, expressed as a percentage of each Member State's GNI, reflecting the minimum accepted level of unlimited financial solidarity between Member States. Net positions exceeding such a threshold will be eligible for a correction (partial refund), thus giving an insurance against excessive net contributions. Conversely, the total volume of corrections (refund volume) will be limited to a maximum amount, thus insuring those not benefiting from a correction against excessive costs of the mechanism. If the sum of all corrections exceeds the total predetermined volume, the refund rate is reduced accordingly.

The proposed generalized correction mechanism will decrease the negative net balances, reduce the spread among net contributors, and, on the other hand, lessen the financing burden of those who do not benefit from the mechanism.

ANNEX I

Criteria for assessing own resources

Visibility and simplicity

Visibility of the financing of the EU to the citizens is essential in order for the public at large to be aware of the cost of financing the Union, whereas simplicity is an important condition for a system to be understood by the public at large or by the interested citizens, thereby facilitating public scrutiny of policies. Indeed, as taxpayers tend to question the use and the amount of the contributions they pay, they also force the authorities to better justify the use of their resources and to make the best use of them. Thus, visibility and simplicity may impact on the accountability and on the overall efficiency of the system.

A tax-based own resource levied on citizens and economic agents to finance the EU budget could increase the visibility of the financing and therefore also awareness of the Union and its cost. Furthermore, an increased reliance on fiscal resources could contribute to a greater involvement of the European Parliament also on the revenue side of the budget, which could have positive consequences in terms of efficiency.

Financial autonomy

In this report financial autonomy is defined as autonomy of the EU budget from national treasuries. Autonomy is enhanced if there is a direct link to citizens, i.e. if the budget derives (at least a significant part of) its resources directly from taxpayers.

A high degree of financial autonomy weakens the foundation for discussions on national budgetary balances vis-à-vis the EU budget, allowing instead for a focus on common concerns and the general benefit of EU policies.

Fiscal resources assigned to the supranational EU level may also lead to stronger public scrutiny of policy-making at this level and to increased direct responsibility of, notably, the European Parliament towards EU citizens. This may, in turn, enhance the democratic accountability of the EU financing system and contribute to closing the gap between EU institutions and EU citizens.

Furthermore, if a tax is directly related to Union policies (e.g. customs duties) there is clearly a rationale for at least part of the tax to accrue directly to the EU budget, in particular if it entails a mismatch between the geographical pattern of collecting the tax and the geographical pattern of the tax burden.

Efficient allocation of resources

Taxes may modify the structure of prices in the economy. This may in turn affect the behaviour of economic agents. In some cases, a change of behaviour is precisely the objective underlying the creation of a tax. This is, in particular, the case where market imperfections or externalities exist, such as in the environmental field. In other cases, such a change of behaviour is not desirable and it can be a source of economic inefficiency. This is, for instance, the case when the tax treatment of a specific investment differs according to its location in the internal market. The ultimate location of the investment may then be determined by tax rather than by productivity concerns.

Hence, the introduction of a tax-based own resource may facilitate the efficient allocation of resources on two grounds. It can increase efficient allocation of economic resources by setting a price-tag on negative externalities when this is not possible on a national level (e.g. cross-border pollution). Furthermore, if the attribution of a resource to the European Union makes it necessary to harmonise a tax base, it can also have potential benefits for the internal market, by contributing to a more optimal allocation of economic resources based on relative productivity and not on differences in taxation levels between Member States. Assessing a tax-based EU own resource requires taking into account these allocation effects.

Sufficiency

As a general principle, it is essential that the resources used for financing the EU be sufficient to cover the expenditures of the EU in the long run. An own resources system should offer the flexibility needed to respond to the evolution of the financial needs of the EU.

If a new fiscal own resource were to replace the VAT and GNI contributions, it would need to be a substantial resource, bringing about revenues equivalent to close to 1% of the EU GNI. Furthermore, should the EU budget increase over time, one would have to make sure that the own resource did not rely upon a base that tended to decrease over time.

At the same time, it is always possible to combine several fiscal own resources to obtain sufficient revenues. However, a combination of too many small resources might result in a system that is too complex. Fiscal own resources can also be combined with resources of a different nature, such as direct contributions from Member States as in the current system. In short, the sufficiency criterion has to be placed within the broader context of the overall financing of the EU budget, rather than at the level of individual own resources.

Cost-effectiveness

The operating costs for individual own resources and for the system as a whole are a matter of interest, both to the economic operators and taxpayers and for the public administrations. Increased co-ordination related to the collection of an EU own resource could under certain circumstances lead to substantial cost-savings for taxpayers and/or administrations of Member States. In other cases, an EU own resource could, on the contrary, be applied in addition to existing taxes, even if the overall tax burden remains unchanged, thus imposing a new administrative burden on taxpayers and/or administrations. It should, however, be noted that a financing system implying higher costs at EU level would not necessarily imply the existence of more inefficiency from a global perspective. Operating costs could decrease particularly in the case of transfer of tax revenue to the EU level that is characterised by so-called 'regional arbitrariness' (see previous section). In this case, complex tax-sharing rules have to be defined, e.g. for customs duties or corporate income taxation, which sometimes prove costly to both taxpayers and tax administrations. In these cases, it may be more efficient to assign the tax to the 'higher' level of authority as, for instance, in the case of customs duties in the EU.

Stability

The need for a balanced budget requires stable resources.

Own resources can bring about more or less stable revenues across time. For example, some revenues may be very sensitive to the business cycle or to the price of commodities. As a consequence, EU revenue may be insufficient for a given year, while it may exceed the needs in the following year. In short, even though the sufficiency criterion might be respected in an average year, short term variations in the European Union's own resource revenues may prove particularly difficult to manage.

However, this is not so relevant for the assessment of an individual resource if the appropriate adjustment mechanism within the EU financing system is ensured by another resource that ensures overall stability of EU revenue. In particular, it may be useful that the financing system includes a residual resource, which automatically adjusts revenue to changes in budgeted expenditure in order to maintain balanced budgets. Alternatively, increased tax autonomy of the Community could also be accompanied by more autonomy in terms of capacity to borrow from (or lend to) the financial markets. This would allow the Community to soften the budgetary impact of exogenous revenue shocks. However, this last, politically sensitive, issue falls beyond the scope of this report and is not dealt with further in what follows.

Equity

The design of a financing system has implications in terms of equity. Equity criteria may refer to the situation of individuals or to the fair burden-sharing among Member States.

Two concepts are usually used when it comes to equity applied to citizens. First, 'horizontal' equity refers to the principle whereby 'equals should benefit from equal treatment'. At European level, this principle has an important symbolic value. Equity between citizens implies a large degree of harmonisation of the relevant tax base and a common tax rate.

The 'vertical' equity criterion focuses on distribution of income among citizens. Vertical equity would in particular require that the burden of the financing system would be consistent with the ability to pay of citizens. Given the relatively small size of the EU budget (about 1% of the Union's GNI) and taking into account the prime responsibility of Member States in relation to redistribution policies, however, vertical equity is not considered here.

Equity between Member States, expressed as proportionality of gross contributions to the national income is an important political criterion. It is, however, less relevant for a tax-based resource levied directly on citizens or economic operators, in particular if it is characterised by 'regional arbitrariness'. Regional arbitrariness refers to a situation where it is difficult to determine the exact share of a tax base to be allocated to individual Member States or where there is a high (potential) mismatch between the country collecting the tax and the country of residence of the economic agents bearing the burden of the tax.

In many multi-tier government organisations, problems related to fairness are dealt with using equalisation mechanism to redistribute revenues. If the burden of a tax-based own resource falls too heavily on taxpayers in one Member State, some form of financial transfer may be organised to compensate that country. The need for such a transfer is closely related to the degree of regional arbitrariness of the fiscal resource.

Notwithstanding the wide recognition of the importance of solidarity between the Member States in the context of the European Union, some budgetary adjustments may be required in order to avoid 'excessive' contributions to the EU budget. This issue is addressed in part II of this report.

ANNEX II

Budgetary balances and correction mechanisms

- As corrections for the year 't' are reimbursed and financed one year later, averages in the tables are calculated over the six-year period 2008-2013 since the 2007 correction is paid in 2008 and the 2013 correction takes place in 2014.

- Tables illustrate the data for the 25 current Member States. Where necessary for the overall coherence the data for Bulgaria and Romania have been included as well.

>REFERENCE A UN GRAPHIQUE>

>REFERENCE A UN GRAPHIQUE>

>REFERENCE A UN GRAPHIQUE>

>REFERENCE A UN GRAPHIQUE>

>REFERENCE A UN GRAPHIQUE>

>REFERENCE A UN GRAPHIQUE>

>REFERENCE A UN GRAPHIQUE>

>REFERENCE A UN GRAPHIQUE>