Lessons Learned by Comparing Federal Systems


2 Colum. J. Eur. L. 583 (1996)

Michael Bothe. Professor of Public and International Law, Johann Wolfgang Goethe-Universität, Frankfurt. Co-director, Institute for Comparative and International Economic Law; Dr. iur. Heidelberg.

At the end of any scientific endeavor like this one, it is always good to come back full circle, to the very beginning. In the beginning, there was, as there should have been, the question – the good question. At the end, one hopes there is an answer. Let me take up the questions George Bermann raised in his introduction and try to find out whether we know more about the answers.

I will start with the “why” of a federal distribution of power and come to the “how,” “who” and “what” as we go along.

The essence of the discussion on the advantages and disadvantages of federalism is concern about good government. But what is “good government”? We could define good government as the capacity to face up to the perceived challenges of a given society. Our idea of good government is, thus, determined by our perception of these challenges, of the problems we want solved. This very easily leads to one fundamental criticism of federal distribution of power: namely that problems in our global village go beyond the scope of single political entities – smaller entities, in particular. At first glance, this would seem to suggest that one should not look to the advantages of decentralization (of giving more power to lower levels of government), but rather to creating new regulatory schemes at higher levels, particularly at the international level. However, depending on the type of problem, experience shows that decisionmaking at lower levels (which is closer to the actual problem) has its own merits, and regulation at higher levels is not necessarily better-suited to solving problems, even those having considerable geographic scope.

As we raise the question of good government to a high level of generality, the concern becomes the general welfare or well-being of people. Two more questions then arise: First, is the promotion of welfare a governmental task at all or should it be left to the invisible hand of the market? Second, whose welfare do we mean, or who is to be the beneficiary of government action? Both questions matter for our subject, because the question of the vertical distribution of regulatory power only arises if and to the extent that there should be regulation at all. A need for regulation is seen particularly where relevant political forces feel that the benefits of non-regulation lead to unacceptable outcomes.

The most common answer to the first question is that no clear-cut choice can be made but that the right mixture must be found. In recent times, there seems to be stronger preference for the invisible hand of the market. But this does notmean that government has become irrelevant. Government must ensure that the invisible hand works properly. Here, we come from the “why” to the “what.”

If the invisible hand is to work properly, an important purpose of government activity is the prevention and removal of barriers to the adequate functioning of the market. This is indeed a matter of international concern, addressed by numerous mechanisms and organizations, such as GATT/WTO on the international level, and the EU, the MERCOSUR or NAFTA on the regional level. But it is also a concern for federal states, one which becomes more important when substantial regulatory powers are left to lower levels of government. This is the reason for such concepts as “undue burden on interstate commerce” in the United States, or the prohibition of measures having an effect equivalent to quantitative restrictions under Article 30 of the EC Treaty.

This then leads to the question of centralized or decentralized enforcement mechanisms. The first type includes ex officio measures of enforcement taken at the higher level of government, such as those provided in Article 169 of the EC Treaty. The second type includes initiatives of the higher level of government which are then implemented by decentralized actors (GATT panels, the federal courts in the United States, and the Court of Justice of the European Communities). The third type is unilateral enforcement – i.e. enforcement through whatever pressure decentralized actors might be able to exert. This is already highly controversial at the international level and completely unacceptable in a more integrated society. In the European system, there are still some traces of this unilateral enforcement, although fortunately only weak ones.

Regulation comes in through the back door (so to speak) where for one reason or another market forces cannot work properly. For example, regulation may be designed either to prevent private parties from restricting competition or to address market failure. Market failure, however, remains to be defined, because what is considered market failure may vary from society to society. Generally speaking, the problem is how to protect social values without unduly affecting the functioning of the market. An important example of this problem is taxation. Raising revenue for any governmental purpose affects the functioning of the market. This is why government spending has come under attack in our systems. But this example clearly shows that we really cannot live without some kind of interference with market forces. Thus the question is: to what extent is this interference legitimate?

If and to the extent that there is a need for uniform regulation, four approaches are possible: federalization, approximation, harmonization or living with decentralized measures. These four possibilities play an important role in EC law – they can be found, for instance, in Articles 100 and 100a of the EC Treaty. The result of the first three options is uniformity, and the difference between them is structural or procedural. The distinction lies in the consensus problem: the type of consensus required is different for the three options. This problem of consensus is fundamental in a federal system, and the more consensus needed, the more difficult it is to find a solution. On the other hand, consensus-based solutions are more acceptable and easier to implement. The basic political question is: How much political consensus is needed on which level of government?

There is a link between the choice of the appropriate level of government and the existence of a consensus. Consensus might be easier to achieve at a lower or higher level of government, as the case may be.

A particular way of achieving what might be called “vertical consensus” is integrated decisionmaking. Majority voting in the EU/EC Council is only the tip of the iceberg. The reality of the decisionmaking process involves the political and bureaucratic interlacing of the Commission and the ministerial bureaucracies of the Member States. A similar situation prevails in Germany, where the power of the Bundesrat, together with a few other, extraneous circumstances, have led to the same kind of political interlacing. In German terms, this is known as Politikverflechtung. This phenomenon is probably not entirely absent in the United States, although decisionmaking at state and federal levels in the U.S. is much more distinct than in Europe or in Germany.

Because we have defined good government as one which is able to face the perceived challenges of a particular time, it follows that it is this perception which matters, whether with regard to the legitimacy of government interference with the invisible hand of the market, or the perceived need for uniformity. This is all, then, a matter of history, of political culture as it develops at a given moment in history. It is thus possible to observe waves of centralization when relevant actors feel a need for new government functions. This explains the centralizing trend of the New Deal era in the U.S. and also certain centralizing effects on earlier developments in environmental policy in the 1970’s. In our time, there is a marked trend towards smaller government, or at least considerable lip service to it, which may naturally imply that decentralization is on the rise. Sometimes, one also detects the higher level of government’s wish to decentralize (unpopular) decisions on spending less money.

The choice of the appropriate level of government may be determined both by considerations of efficiency and of social acceptability. It is difficult to see how intellectual property could be efficiently protected if the federal level were unable to legislate in this area. Both Professors Ginsburg and L6wenheim seem to agree on this, although a note of caution is added by Dean Haemmerli. Also, an efficient fight against a “race to the bottom” in certain areas of law is impossible unless federal powers exist in relation to the particular questions raised. Professor Briffault shows that the states, if they were not forced by federal requirements to uphold a certain level of welfare benefits, were rather inclined to yield to the pressure of interstate competition by lowering welfare spending, and Professor Alstott vividly describes the current United States debate on this issue. Financial resources may be raised more efficiently if taxing legislation and administration take place at a higher level of government.