VoxEU Column

Subsidiarity: The forgotten concept at the core of Europe’s existential crisis

In this column, Jean-Pierre Danthine, a co-author of "Making Sense of Subsidiarity: How Much Centralization for Europe?", revisits the report nearly 25 years on from its publication. He examines the main themes of the report and shows how such areas as centralisation/decentralisation, subsidiarity, and macroeconomic stabilisation have played out over the years since the report was published. He concludes that the report was both prescient and, at the same time, represents a view from the past of the 'road not taken'.

Preparing for the EU’s 60th anniversary, the European Commission presented a White Paper that outlines five possible avenues for EU institution future (European Commission 2017). This  is not a new problem. The basic economic, political and institutional issues were laid out and analysed in a 1993 CEPR report authored by David Begg, Jacques Crémer, Jean-Pierre Danthine, Jeremy Edwards, Vittorio Grilli, Damien Neven, Paul Seabright, Hans-Werner Sinn, Anthony Venables and Charles Wyplosz (Begg et al. 1993). 

Download the publication for free here (create a VoxAccount if you don't already have one)

Almost a quarter of century later, with Europe in the middle of an existential crisis, this 165 page report, Making Sense of Subsidiarity: How Much Centralization for Europe?, is well worth reading (if I may say so myself). Indeed, the disenchantment with Europe can arguably be traced to the failed application of the subsidiarity principle that was enshrined in the Maastricht Treaty. 

Today, as in the 1990s, Europe’s deep-seated institutional design problem is tied to the inevitable trade-off between efficiency-enhancing centralisation and democracy-enhancing sovereignty. If the trade-off that Europe has chosen cannot be explained and justified to citizens, consequences are unavoidable. The loss of sovereignty can easily turn into a loss of identification with the European project, and, as seen today, the missing identification can generate a dangerous democratic deficit, and a concomitant longing for the autonomy of the nation-state. 

In light of the several options considered in the ‘White paper on the Future of Europe’, the discussion on subsidiarity outlines one direction for reconciling the European project with what is, 24 years later, a surprisingly strong appetite for national sovereignty.

As the authors of Making Sense of Subsidiarity proclaimed, centralisation can often be justified. “In many fields - competition policy, industrial restructuring, some kinds of environmental regulation, for example - application of the principles we have analysed makes us willing, even enthusiastic centralisers.”  But in many other areas it cannot be justified and in those cases the failure to take literally the principle of subsidiarity appears to have had very negative consequences. In essence, the 1993 report details the route that was not followed. In several instances, the report is fascinatingly prescient. At another level, the quarter of century since the writing of the Report makes for a fascinating historical episode. 

A statement on the burden of proof 

Making Sense of Subsidiarity starts with the observation that the subsidiarity principle introduced in the Treaty of Maastricht lacked precision and content. It is “a general political principle rather than a source of explicit guidance”. This important concept risked being buried under the centralization culture of many European politicians and the Leviathan tendencies of Brussels. This Leviathan-leaning was viewed as all the more damaging because “the existing mechanisms for allocating power between the Community and its member states are surprisingly unclear and informal and they do not appear to rest upon a compelling economic or legal logic.” 

The Report noted in particular that “the Community has already acquired the decision-making rules appropriate to a federal state. Without, however, possessing the corresponding political institutions that have historically accompanied such powers in the development of other federal states. For example, in contrast with most federal states, the EU has no formal mechanism for the allocation of competences within the Community outside the areas specifically dealt with in the Treaty. The Treaty itself neither provides a full list of areas in which the Community may exercise competence, nor even lays down the principles according to which competences maybe allocated in the future.”

As we argued in 1993, the principle of subsidiarity is really a statement of who bears the “burden of proof”. Specifically, subsidiarity means that proponents of centralisation are the ones who have to prove that further integration is justified. If they fail to make the case, subsidiarity means that the powers should remain de-centralised. That is, the default outcome is for power to be exercised at more local jurisdictions unless convincing reasons can be found for assigning them to more central jurisdictions. As the report stated, this is particularly important because “the burden of proof will affect not just the allocation of power today (which may matter little), but also the direction in which it evolves in the future (which may matter very much, even if it is hard to anticipate.”

Centralisation, decentralisation and credibility

The report outlined the main merits of centralisation and decentralisation. Centralisation is likely to be desirable in the presence of two simultaneous failures of decentralisation: 

  • First, that non-cooperative policy-making yields results that are significantly worse than cooperative policy-making; and 
  • Second, that agreements to cooperate without centralising are not very credible. 

Centralisation also involves costs, notably in the risk of diminished accountability. 

Decentralisation, on the other hand, increases the credibility of the state's commitment to local differentiation of policies. In sum, “centralisation should in principle increase the credibility of cooperation between localities, and decentralisation should increase the credibility of the accountability of the state to the needs of localities.” 

On the face of it, there is no reason why either of these desiderata should take precedence over the other. The principle of subsidiarity claims, however, that when in doubt, decentralisation should be preferred. “It can therefore be interpreted as the expression of an essentially political judgement that good government is more likely to be under threat from failures of accountability than from failures of cooperation, and moreover that the kinds of distortion induced by these failures in accountability are of the kind that decentralisation can help to alleviate.”

As the 1993 Report notes, “accountability is a notion that economics has often ignored, not because it is unimportant but because it is hard to analyse systematically. (But) … progress in analysing accountability is possible, and the growing scholarly and public awareness in recent years of the nature of government failure and the weaknesses of public choice mechanisms is a good place to start. It is particularly important to do so, otherwise we risk allowing the benefits of centralisation (which often arise due to spillovers and other phenomena that can be quantified) to drive policy conclusions without due attention to the costs, which are typically harder to measure.”

The case-by-case analysis of the merits of centralisation and decentralisation reaches its clearest and most provocative (and they remained so today) conclusions in two areas: social policies and regulation.  

Failure to apply subsidiarity: The Social Chapter 

After a careful review of the legitimacy of the fears of social dumping, the Report concluded that it found “no reasons to believe that the process of economic integration will undermine the ability of individual member states to make optimal decisions on social and health regulations in accord with their national preferences. If such regulations have been set too high or too low due to dysfunctional national labour markets, the integration process will help by decreasing the market power of unions, insiders or employers as the case may be and it will erode the associated rents. There is thus no case for centralizing or even coordinating social policies. The Social Chapter of the Maastricht Treaty is in fact in direct contradiction with the subsidiarity principle that the same Treaty espouses.”

By contrast, the Report views regulatory harmonisation as one where the centralisation argument makes sense.  “In many regulatory areas - with the notable exception of drinking water regulation - we see significant merit in centralised or partially centralised policies, and in many respects the Community's current practices are well thought out.”

Misapplied subsidiarity: Agricultural and regional policy

One of the Report’s striking and controversial conclusions is that two of the areas in which the case for centralisation of EC power is weakest are the two which currently represent the largest components of the Community's budget: agricultural spending and the Social and Regional Funds. 
Looking to the future, these observations suggest that the principle of subsidiarity – if taken seriously – warrants a major rethinking of the Community's current spending priorities. As we wrote in 1993: “The case for centralisation appears to be strongest in areas whose budgetary requirements are not high and weakest in those where the budgetary cost is large.” That these remarks remain timely and spending priorities have not been altered can of course be viewed as an indication that the call for taking seriously the principle of subsidiarity indeed has failed to be answered. 

Fiscal policy and macroeconomic stabilisation

By contrast, the chapter on fiscal policy coordination illustrates how deeply things have changed in the related areas of the EU. This is the case notably because the analysis was performed before the advent of the euro and before the Global Crisis. The discussion thus takes place in what is a fairly abstract level viewed from today, or in a sense, under the veil of ignorance. The Report concludes that: “On balance, we cannot make a convincing case for centralising fiscal policy for macroeconomic stabilisation. Even though EC-level insurance is appealing, too many drawbacks undermine that case. Insurance is unappealing because it generates perverse behaviour and attracts only the higher-risk countries.” 

The Report noted the dilemma facing the Community. “The Maastricht limit on borrowing threatens to set up pressures to centralise fiscal policy because national policy is no longer adequate to the task, as we saw from the pervasive failure to use fiscal policy to cope with the recession of 1992-3. Embarking on fiscal centralisation is likely to unleash Leviathan pressures and weaken accountability.” Thus “the Maastricht Treaty is internally inconsistent. It calls for subsidiarity which would discourage the emergence of active EC-level fiscal stabilisation; yet by establishing very tough ‘prudential rules’, it will act as a powerful inventive to break the subsidiarity principle. As a corollary, if fiscal policy should remain at the national level, the Maastricht Treaty’s provision must be either suppressed or implemented with considerable flexibility.” 

When we wrote it, we went on to forecast optimistically that “If or when monetary union is attained, an independent central bank provides the appropriate commitment to price stability; knowing that they are unable to print money to finance deficits, member states will then have to pursue more responsible fiscal policies.” (my emphasis). Of course this did not play out as hoped. Indeed, the general explosion of public debts and the fact of monetary Union require an update of the analysis of the merits of centralisation in this area. 

Factor mobility, fiscal competition and the survival of the nation state 

The prediction of increased factor mobility and the impact of the resulting fiscal competition are the focus of an important chapter of the Report. The authors note that “although inertia may be sufficient to permit moderate redistribution between different types of factors that are not de facto very mobile, the feasible extent of redistribution within countries is curtailed by an increase in the mobility of factors, notably of capital and the most affluent workers.” That this likely evolution presents a significant challenge is clearly recognised. Yet, “the basis of the principle of subsidiarity is the presumption that government failure at the centre remains of pressing concern. Pointing to potential deficiencies of fiscal competition is not in itself sufficient to legitimate the growth of a federal fiscal structure in Brussels that risks capture by rent-seekers and faces difficulties in remaining accountable to the citizens of the EC.”

It is here, however, that the Report is most sanguine on the potential (central) role for Brussels. “Europe's fiscal future cannot be built on the principle of exempting from the burden of financing public expenditure those who happen to be able to escape national tax collectors. Escape routes must be closed or at least curtailed. Accountability, distributive justice and economic efficiency require that action be taken. Where mobility of products or factors raises acute difficulties, the dangers of collective action are outweighed by the certain consequences of collective inaction.” In trying to follow the precept: 'decentralise where possible, coordinate where necessary, centralise only when that coordination would not be credible,’ the Report advocates some changes in the tax system as the appropriate response in light of subsidiarity. Thus given that “the erosion of capital income taxes in a Europe of fiscal competition is a safe prediction” with important implications for income and wealth distribution, a move to cash-flow taxes, by avoiding the need to centralise taxation of capital incomes, would permit “making use of the principle of subsidiarity to pursue accountability through government at a lower level.” In the same vein, the imposition of lower bounds on the national VAT rates is the correct response to detrimental tax competition at this level.

But protecting the Welfare State may well be one of the main role of centralisation at the EU level. “Centralising aspects of the insurance or welfare state would overcome those particular difficulties that arise from mobility among lower-level jurisdictions to avoid taxation or take advantage of generous welfare systems.”

The Report notes, however, with prescience, that “centralisation may run into another difficulty, namely that people in the UK may be unwilling to contribute to support the poor in Greece, or people in Germany to support people in Portugal. When such problems arise, there will be a limit to the extent to which any form of coordination at EC level can mitigate the erosion, albeit partial, of the welfare state that will be inevitable under decentralisation.” 

Today we know that attempts to go beyond what people are willing to accept in this or other areas may lead to the radical decision to exit the EU, a problem of time inconsistency that the authors had invoked as one of the limitations of a centralised response to the macroeconomic stabilisation problem: “A Community insurance system must face the possibility that a country - or worse, a group of countries – may be subject to a particularly bad and long-lasting shock, possibly a permanent one. If the costs are high, the incentive to renege may be overwhelming. Not only would the insurance system collapse but its demise could very well generate deep misgivings about other areas of common interest.”


Begg, David, Jacques Crémer, Jean-Pierre Danthine, Jeremy Edwards, Vittorio Grilli, Damien Neven, Paul Seabright, Hans-Werner Sinn, Anthony Venables and Charles Wyplosz (1993), Making Sense of Subsidiarity: How Much Centralization for Europe?, CEPR, London

European Commission (2017)

5,984 Reads