2. Fiscal Federalism and the European Union
If one re-thinks the EU budget with an open mind, then the right question should be, “what for”? What is the optimal distribution of functions of government among the different levels, European, national, regional/local? The theory of fiscal federalism can help a great deal in finding the right answers.
As developed by Richard A. Musgrave in 1959 and by Wallace E. Oates later on,23 it is based on three main economic functions of government: stabilization, distribution and allocation. According to this theory, stabilization and distribution are better assigned to the central level of government. Allocation can instead be subdivided across the different levels of governments following a criterion of coincidence between beneficiaries and taxpayers.
What matters is the spatial incidence of benefits: those who benefit from the supply of goods at a certain level of government should be the same who bear the burden of financing it. Thus, functions such as national security (defense), external representation (diplomacy) and justice should be carried out by the central government, while transportation and sanitation, say, can be assigned to lower levels of government.
Macroeconomic stabilization is also and quite obviously for the central level of government: not only the money supply and the exchange rate, but also counter-cyclical fiscal policy. Similarly, income re-distribution is best carried out at the central level.
A division of labour among different levels of government in the supply of public goods is justified by the territorially diverging preferences within a large jurisdiction such as a federation. Some goods and services provide territorially circumscribed benefits. In these cases, a de-centralized supply allows local jurisdictions to provide quantities that correspond to specific costs and preferences.
As a corollary, taxation of highly mobile factors, such as capital gains and to a lesser extent profits, should be assigned mainly to the central level of government. Labour is also mobile, albeit to a more limited extent. Its taxation is linked to social security, a function of government carried out at federal level in the U.S. and at member state level in the EU. Consumption taxes as well as estate taxes, on the contrary, are best left to lower levels of government.
The theory of fiscal federalism has always been applied to existing federal states, but there are several reasons why it can tell us something useful also about the EU. First, even though the EU in its present form is not a federation, the least that can be said is that it is nonetheless a specific level of government for Europe as a whole – arguably the highest as testified, among other things, by the precedence community law has with respect to member states’ laws.
Second, the EU contributes to Europe’s macroeconomic stabilization via its monetary policy and via its admittedly limited spending policy24. Third, it complements member states in carrying out a redistributive function via the CAP, the structural funds and other programs from its budget.
Fourth, it complements member states in the provision of some public goods that fiscal federalism typically assign to the central level of government: defence, particularly peace-keeping; diplomacy, particularly development and humanitarian aid; scientific research25.
Finally, it’s an “ever closer union”, i.e. a jurisdiction in flux since its creation that may be tasked with new functions of government, may remain the same or even be downsized – all cases in which fiscal federalism has also a lot to say on how to optimize its taxing and spending.
A real world example of what a federal budget responding to the precepts of fiscal federalism may look like is the U.S. federal budget, which is sketched below as a percentage of U.S. GDP (the year is 2007). It is presented here mainly as a benchmark.
% of U.S. GDP
– Defence 4.0
– Social Security 4.3
– Medicare and Medicaid 4.9
– Unemployment and Welfare 2.6
– Interest on the Debt 1.7
– Everything else 3.2
When it comes to social security, particularly pensions, there coexist in Europe many different national preferences, due to different demographic trends and to varying degrees of influence on the part of different interest groups. Many of these groups would actively oppose the transfer to Brussels of this function of government.
On top of these collective action hurdles, national pension systems in Europe are mostly based on “pay as you go” (active workers’ contributions are used to pay retirees) and have such diverging degrees of fiscal liabilities as to make the change over a centralized system next to impossible from a public finance point of view. The transition would, in other words, be truly intractable in today’s Europe even for iron-willed Martians – even though a European wide social security system would greatly help labour mobility within the internal market.
Welfare (including health care) and income re-distribution to individuals and households would also be best left to member states in today’s Europe, even though, as with social security, this clashes with fiscal federalism precepts. The reasons are in fact similar: national preferences and traditions vary across the continent, member states have hugely different degrees of generosity/expectation in the supply/demand of this particular public good and their convergence may still require a lot of time.
On the other hand, the theory of fiscal federalism may have its best ally in the principle of subsidiarity, enshrined in the Treaty itself. It reads: “the Community shall take action […] only if and in so far as the objectives of the proposed action cannot be sufficiently achieved by the Member States and can therefore, by reason of the scale or effects of the proposed action, be better achieved by the Community.”26
Although the principle of subsidiarity is quoted more often by those who think that the Union’s competences are already too many and be better circumscribed, the letter of it leaves the door open to a cold-minded reappraisal of who should do what in Europe. In particular, the words “by reason of scale or effects” evoke the importance given by fiscal federalism to externalities and spillover effects.
Thus, even excluding big ticket items such as social security and welfare, there still remain other functions of government that according to fiscal federalism and a Martian’s common sense, would be best assigned to the central, EU level of government. These are:
security and defence;
diplomacy and foreign policy (including development and humanitarian aid);
border control (the equivalent of Homeland Security in the U.S.);
infrastructural projects with European-wide network effects;
large-scale Research & Development projects;
The last item, regional re-distribution, would be kept for as long as needed to help the countries that joined the EU in 2004-2007, and others that may join in the future, to close the gap in living standard with the rest of Europe. Resources under this heading would go to richer member states only occasionally, to counter specific emergencies such as the current one in Greek public finances.
Since the provisions of all these functions of government would entail a substantially bigger budget relative to the European economy as a whole than it is the case now, the kind of macro-economic stabilization at federal level called for above should become possible. Brussels would finally have the ability to transfer a sizeable amount of resources from states doing better to states doing worse taxing the former more, and spending more in the latter.
The provision of other public goods would be best left where it largely is now, at member state level, in order to satisfy the different underlying preferences. Examples include education, culture, health, transport etc. In these and other cases the EU should simply divest itself of what it does today, except regulating through an internal market mandate, or coordinating national actions when needed.
To give an example: its regulatory powers on the food chain enable the Commission to legislate on veterinary controls and monitor their implementation on the part of member states, even though this, if seen as a public health issue, would not fall within the competences fiscal federalism would assign to the central level of government.
On the contrary, education and culture are best left entirely to member states and even the relative small amount of resources from the current EU budget spent on them are wasted and should be eliminated altogether. Doing little things here and there, even outside its core and most logical competences, is precisely what makes the EU vulnerable to the charge of “creeping federalism” – and, worse, inefficiency.
The current budget of the EU includes some funds for scientific research. R&D spending in the 27 EU members was € 212.8 billion in 2006, or 1.84% of their combined GDP5. To this the EU budget added less than € 4 billion, or 0.03% of GDP.27 In other words, on any given euro spent on R&D in the EU, 98 cents come from member states and 2 cents from Brussels. On top of that, these two cents cannot be concentrated anywhere or focussed on anything, but must be spread far and wide among 27 member states and tens of different projects.
Nonetheless, these funds are very much sought after, further contributing to a public perception of the EU as a cash cow. This happens partly because there are in fact very few American-style foundations in Europe that support cultural, scientific and social activities. And partly because “showing the blue flag” of being EU-funded, even if for trivial amounts, is often considered prestigious, especially in Southern Europe or in the new member states.
It does not take an extra-terrestrial point of view to understand that it would make much more sense to put the EU in charge of truly pan-European scientific projects on a scale beyond what even the biggest member states can now afford on their own. Here the logical parallel is with the U.S. and its space programs or its national security-related research efforts – all run at the federal level. Having the EU level of government take care of defence would, of course, greatly facilitate running ambitious space, aeronautics and defence-related research projects on a European scale.
The transfer of certain functions of government from the national to the European level should entail no net increase of public expenditures in the EU as a whole, and possibly a net decrease, due in fact to economies of scale. To remain in the defence field, a single organization is likely to come up with a “bigger bang for the buck” than 27 different ones. On top of that, as the NATO cold war experience shows, efforts to coordinate independent defence establishments have always produced disappointing results and lots of free riding at the expense of the richest suppliers of the public good.
Diplomacy, a relative inexpensive item in the budgets of present day nation states, would nonetheless cost a lot less if centralized at the European level. The European commission already has delegations in more than a hundred capitals, the head of the delegation often accorded the rank of ambassador. Member states embassies around the world that would become redundant are in the order of thousands. Simple arithmetic tells us that only the intra-UE embassies are in the order of several hundreds (26×27=702). Some of the biggest member states’ embassies are in other members states: Italy’s in Berlin, France’s in Rome etc.
Development aid and humanitarian aid – things that the European Commission already does on a substantial scale (it is the first world donor of the latter aid, for example) – would still be done, but not only on even larger scale, but also and for the first time in coherence with the overall goals of an EU foreign policy. Today, the lack of this – as opposed to the very existence of 27 distinct member states’ foreign policies – has led the EU to disburse its humanitarian aid under the absurd and hypocritical assumption, particularly in situations of conflict, of being politically neutral.
The European Union is first of all a customs union. Customs is, thus, an exclusive EU competence28: member states have no longer any intra-EU border controls, share a common external border, and implement the same customs code, together with other regulations concerning for example the control of export of dual use goods or UN resolutions.
Despite this, member states still anachronistically rely on 27 separate national customs organizations. Given also the growing importance of new security missions (anti-proliferation, anti-terrorism, health and food security etc.) at the expense of the traditional tax collecting mission, EU customs is a prime candidate to integration at the central level of government on a model that may very well follow the U.S. Department of Homeland Security.
Agricultural subsidies represent 43% of the current budget of the EU and are its biggest single item. There is an immense literature harshly criticizing them for their negative effects on resource allocation, domestic and international prices, world trade, food quality and the environment. Their continued existence is testimony to the lobbying strength of European farmers and to the explicative strength of Mancur Olson’s logic of collective action.
As a Martian, however, I ignore the farm lobby and design an EU budget with zero resources for the Common Agriculture Policy (CAP). Note that doing this would not necessarily entail the death of the CAP. Reforms already undertaken are shifting in any case the emphasis from subsidies to production and price fixing to the direct support of farm income. To the extent this support does not create market distortions, it can be then carried out by the single member states according to their own national preferences, leaving for the European Commission a role as coordinator of, and watchdog on, commonly agreed rules – as it already happens in a number of policy fields.
Summing up, I have tried to apply some insights of the theory of fiscal federalism to the distribution of competences among the different levels of government in the EU. I did this clearly and unabashedly in a highly subjective way, sometimes following and sometimes disregarding the precepts of the theory.
Leaving welfare and social security at member states’ level, while moving defence and diplomacy and the federal, EU level, may strike the reader as strange, for example. But as a matter of fact, I consider the creation almost from scratch of a European army and a European diplomatic corps in the world that resulted from the end of the cold war easier than, for example, finding a formula to merge into one the 27 different retirement systems and pension provisions today in force in Europe.
Nick Witney, the former head of the European Defence Agency, wrote the most knowledgeable and persuasive indictment of Europe’s security and defence policy as it stands today.
“Nearly two decades after the end of the Cold War – he wrote – most European armies are still geared towards and all-out warfare on the inner German border rather than keeping the peace in Chad or supporting security and development in Afghanistan […] This failure to modernize means that much of the € 200 billion that Europe spends on defence each year is simply wasted […] The EU’s individual Member States, even France and Britain, have lost and will never regain the ability to finance all the necessary new capabilities by themselves”.29
If this is the diagnosis, and if several years spent at improving coordination and cooperation among different national defence organizations have failed as a cure, wouldn’t the creation of a EU army be the most logical step to pursue? Note that precisely because the mission of Europe’s military force has so profoundly changed, it is in principle much easier to start a new army from scratch – manpower, equipment, doctrines and all – rather than persevering in the futile attempt to convert existing ones to new missions trying at the same time to increase their cooperation. Why is it possible to create from scratch a new currency and a new central bank and not a new army?
The distribution of competences across the different levels of government is not an exercise that can be done following mechanical rules. It rather needs to be adapted to historical, political and economic circumstances and ultimately supported by popular consensus. On the other hand, weighing these circumstances is inescapably subjective and public opinion can sway and change. Rarely what one day looks politically unfeasible stays that way the next.
Let’s now look, first, at what the current EU budget looks like as a percentage of Europe’s GDP.30
% of EU GDP
– Agricultural subsidies 0.43
– Subsidies to poor regions (structural funds) 0.35
– Internal policies 0.10
– External policies 0.06
– Administration 0.06
Again, in order to provide a benchmark, I reproduce below the main items of the U.S. Federal Budget as a percentage of U.S. GDP (2007).
% of U.S. GDP
– Defence 4.0
– Social Security 4.3
– Medicare and Medicaid 4.9
– Unemployment and Welfare 2.6
– Interest on the Debt 1.7
– Everything else 3.2
And finally these are the EU functions of government I have been sketching here, together with a ballpark quantification of their cost as a percentage of Europe’s GDP.
% of EU GDP
– Defence 1.0-1.5
– Homeland Security 0.3-0.5
– Research and Development 0.5-1.0
– Trans-European Networks (TENs) 0.3-0.5
– Regional re-distribution 0.3-0.5
– Development and Humanitarian Aid 0.5-0.7
– Administration (incl. Diplomatic corps) 0.1-0.3
Why these figures?
Currently, EU member states spend on defence anything from Ireland 0.5% of GDP to Greece 3.8%. Figures for big member states are as follows: France 2.4%, Germany 1.3%, Italy 1.8%, Poland 2.0%, Spain 1.2%, the UK 2.6%. A hypothetical EU defence spending in the order of 1.0-1.5% of GDP may thus seem low. But even the lowest figure, corresponding to some € 130 billion, would automatically make of the EU armed forces a formidable fighting machine, second only to the United States and with 3 to 5 times the resources available to military powers such as Russia, China or Japan.
In particular during the transitional phase necessary to complete the creation of an EU standing army, member states would keep portions of their armed forces and finance them on their own resources – and this would bring up Europe’s overall level of military spending. After the transitional phase, member states would be free to maintain – again on their own national budgets – forces for territorial defence or other duties, in analogy to the U.S. National Guards.
The resources available to the U.S. Department of Homeland Security corresponded in 2007 to about 0.3% of the U.S. GDP. The territory of the European Union has longer and more problematic borders, more ports and airports, more residents to protect. Thus a figure of 0.3.-0.5 % of GDP looks reasonable.
As for R&D, Europe today spends slightly less than 2% of GDP on it, 98% from member states’ sources, public and private. One major goal of the Lisbon strategy on growth and jobs agreed upon by member states in 2000 and confirmed ever since is to reach a figure of 3% of GDP. It seems logic to assume that the gap be filled by the central, federal level of government as argued here.
TENs would get in the hypothetical budget far more than what they get from the present EU budget. But this new level of resources would in part substitute for, instead of coming on top of, what members states now spend at national level to increase the connectivity and interoperability of European transport, energy and telecommunications networks, receiving from Brussels little money and much coordination. Additionally, this is perhaps the most effective way of stimulating the EU economy from a central budget in times of crisis.
The order of magnitude envisaged here for regional cohesion is similar to that of the current structural funds. The only true difference is that the re-distribution effort would be focussed on the new member states and go to the richer ones only in emergencies.
The highest end of the figures suggested for development/humanitarian aid corresponds to the international commitment taken by the rich world at the UN Assembly in 1970 to devote 0.7% of their GDP to development and confirmed more recently in the UN millennium goals. Three EU member states – Luxembourg, Denmark and Sweden – are above that targets, but the rest are way below, so much so that even 0.5% of the EU GDP, or some € 65 billion, even inclusive of humanitarian aid (which is of a different, short term, nature) would be a great leap forward from the current (2008) figure of € 48.6 billion (European Commission plus member states, excluding humanitarian aid).
Lastly, administrative expenses would absorb between two and six times what they do now, but they would support a much larger overall budget, armed forces and a whole diplomatic corps with its infrastructure of embassies and representative offices around the world.
Even though public expenditures have a tendency to constantly grow and seldom contract, I believe the figure – 3-5% of Europe’s GDP – put forward for the functions of governments here assigned to the EU is realistic and perhaps also downwardly flexible: regional cohesion is supposed to be limited in time and the same may apply, hopefully, to development aid; TENs have physical limits; armed forces exclusively conceived for peace-keeping and stabilization missions can be very effective even well within the lower boundary of spending (1.0 %).
On the other hand, both the nature and the volume of this suggested EU budget are apt to be used to stimulate and stabilize Europe’s economy if need be, largely solving the chacun pour soi problems experienced in the 2008-2009 downturn mentioned in the first chapter.
Finally, these new federal functions of governments and the attendant expenditures would be largely substitutive, rather than additional, to national ones – even though a large degree of discretion would be left to member states. For example, the latter would be free to decide the size and cost of their own territorial defence/national guard, as much as whether they prefer to keep some offices of representation abroad (or at fellow member states’ capitals), or support particular humanitarian or development aid programs – not to mention decide on their own the level or public and private resources they want to channel to basic and applied science.
To give an idea of how small the EU budget sketched here is (hardly the stuff of a super-state!), the table below shows the current size of European total general government expenditures, including in some selected member states, in the U.S. and in Japan. Again, please bear in mind that EU spending could very well replace an equivalent portion of current public expenditures – unless, that is, single member states decided otherwise.
% of GDP (year 2007)
United States 37.4
Sources: Eurostat; OECD
I call what emerges from the exercise carried out in this paper “Federation Light”. The term “Federation” is indeed the aptest to describe an entity made up of existing European nation states but in possession of most attributes of state sovereignty: a directly elected parliament, a court of justice, a currency, a Treasury, armed forces, a diplomatic corps, an external border to police,.
On the other hand, it is of some importance to realize that such entity need not be the “super-state” dreaded by Euro-skeptics – at least from the point of view of the relative amount of economic resources controlled by the center, between 3 and 5% of Europe’s GDP.
A federation, yes. But a Federation Light.
The questions left unanswered are countless and also, honestly, beyond the scope of this paper in its present form. I can anticipate just a few.
One thing I completely ignored is, for example, the revenue side of the budget. Here, I can only say that a Federation, no matter how Light, is finally entitled to its own revenues. In other words the Union could at last do away with the current system of transfers from member states – transfers nonetheless, even when they are called “own resources” – and rely upon its own taxes directly accruing to its own coffins. I note only in passing that the resources collected via the TVA throughout Europe are more than enough to finance the budget sketched out here.
Then there are more specific questions pertaining to some specific functions. Can one send an army into battle by majority voting? And the answer is, in my opinion, yes one can and one should. Better still through a vote of the Parliament. According to the U.S. constitution, the Congress authorizes hostilities and declares war, a prerogative eroded by the 1973 War Power Resolution whose reform many advocate.31
On top of that, an EU army would have no nuclear missile to be launched on warning in a matter of minutes but just conventional power to be projected far away from Europe – there would be, in other words, ample time to discuss the wisdom and the merits of any single force projection.
The most difficult question, though, is the political feasibility of this Martian’s view of the EU: it looks scant today and the exercise done here may also be dubbed “thinking the unthinkable”32. Political opinions are not set in stone, however. They depend on rapidly mutable fashions and, more importantly, are subject to change under stress.
The financial crisis of 2008-present has provided plenty of stress to the internal market and the monetary union. European politicians and opinion makers will, I believe, eventually come around to Paul Kugman’s opinion that “[…] the only way out is forward. To make the Euro work, Europe needs to move much further toward political union […]”.
Then we’ll suddenly discover one more time that the unthinkable is actually so perfectly thinkable that it was already thought.
In 1977, a Report to the European commission by a Study Group chaired by Donald MacDougall – who was at that time the Chief Economic Adviser of the Confederation of the British Industry – on the role of Public Finance in European Integration thus concluded33.
“It is possible to conceive, presumably at some distant date, a Federation in Europe in which federal public expenditure is around 20 – 25% of gross product as in the U.S.A. and the Federal Republic of Germany. An earlier stage would be a federation with a much smaller federal expenditure of the order of 5 – 7% of gross product, or roughly 7 ½ – 10 % if defence were included. An essential characteristic of such a federation would be that the supply of social and welfare services would nearly all remain at the national level” .
Note how uncontroversial the use of the term “federal” or “federation” was at the time if it was used in an expert group chaired by a British person. And note also the economic costs of defence (2 ½ – 3 % of GDP) during the cold war.
As for the rest of the quotation, let me just hope that it looks familiar and reasonable by now to the reader of this paper.
23 – See Richard A. Musgrave, The Theory of public finance (New York: McGraw-Hill, 1959). Of Wallace E. Oates, see “An Essay on Fiscal Federalism”, Journal of Economic Literature, September 1999.
24 – At the end of 2008, the EU launched the European Economic Recovery Plan, a stimulus package equivalent to about 1.5% the Union’s GNP. About one fifth of it, or € 30 billion, were resources coming from the EU budget.
25 – The Lisbon Treaty has equipped the EU with: a High Representative for Foreign Affairs and Security Policy (sort of Foreign Minister), currently Ms. Catherine Ashton; a European External Action Service (sort of diplomatic corps); a European Defence Agency (sort of common arms procurement agency). The latter began functioning well before the entry into force of the treaty in December 2009.
26 – See art. 5 of the (Lisbon) Treaty on the European Union. The principle of subsidiarity applies to the many areas in which competences are shared between the Union and member states. Member states can, of course, confer upon the Union whole new competences through further treaty changes.
27 – Eurostat News Release 34/2008, 10 March 2008.
28 – See art. 3 of the “Treaty on the functioning of the European Union” (Lisbon Treaty).
29 – N. Witney, “Re-energising Europe’s Security and Defence Policy”, European Council on Foreign Relations, July 2008, available at www.ecfr.eu.
30 – With small roundings, these figures represent the expenditures (commitment appropriations) foreseen in the EU Financial perspective 2007-2013.
31 – See “Reclaiming the War Power”, Ch. 10 of Cato Handbook for Policy Makers (Washington DC: The Cato Institute, 2009).
32 – Thinking about the unthinkable” is the title of a 1962 book by the American physicist and RAND Corporation strategist Herman Kahn on how a nuclear war could be fought and won.
33 – The full text of the report plus several individual contributions, including one by Wallace E. Oates, who was a member of the Study Group, can be found on the website of the Directorate General for Economic and Financial Affairs of the European Commission at the following address. http://ec.europa.eu/economy_finance/emu_history/documentation/documentation_chapter8.htm