Industrial Policy in Europe—New Options
Paper for Eurokolleg Series, Friedrich-Ebert-Foundation, Bonn
Industrial policy is a highly controversial issue—not only in terms of pros and cons, but also in terms of what actually constitutes industrial policy. Taking a broad perspective, almost any policy has some impact on industry and has therefore an industrial policy content—for instance child care policy to the extent that it influences the decision on the use of either traditional or disposable diapers, and thus discriminates against either the pulp or the cotton industry. Adopting a broad definition, the Single Market project is, of course, by far the most important industrial policy initiative that has been pursued over the last 15 years. Analytically, however, it makes little sense to define the term that broadly. For the purpose of this paper, industrial policy encompasses all those policy measures that are deliberately meant to shape industrial structures.
In view of this definition, many economists tend to argue that there should be no industrial policy at all since it distorts market processes and thus leads to welfare losses; and many of those who do not reject industrial policy in principle are sceptical as to whether the European Union (EU) should play a role in this arena.
The case for industrial policy is based on the observation that markets may fail to stimulate firms to behave in a way that would be beneficial to economy as a whole (market failure). There are different cases of market failure (Dunning 1993, p. 79 f):
. These may present a case for industrial policy if in a given industry the economies of scale are high, thus deterring the entry of initially small firms. Economies of scale generally occur in production but may also be present in research and development (for instance in telecom switching equipment where R&D investment below a certain amount does not lead anywhere). Industrial policy can reduce barriers to entry by giving subsidized credit or facilities to entrants, or by guaranteeing a certain amount of demand for new firms via State purchasing policy. In other cases of barriers to entry, for instance if incumbents are setting prices below costs in order to keep entrants out, it is rather a case for conventional anti-trust policy.
. There is no obvious reason why governments should be better positioned than firms in evaluating the perspectives of, say, technological developments. Therefore, governments' choices between different technological options are likely to be misguided. Such choices should therefore be left to the market.
It is important to point out that the discussion is not on whether or not to introduce European industrial policy. In fact, industrial policy, albeit of a limited scale, has been an important field of activity since the beginnings of the European Communities. What is today called the European Union (EU) has formerly been the European Community which was made up by three entities: The European Economic Community (EEC), its main objectives being the reduction of intra-European trade barriers and the common agricultural policy; the European Community for Coal and Steel, which was set up in the 1950s to manage structural change in these smokestack industries; and the European Nuclear Organisation (Euratom) which was also set up in the 1950s to shape a common nuclear industry (but failed early). With the second and the third entity, the European unification project has included an industrial policy component from the onset, but it has been limited to those three industrial sectors. It was only in the 1980s that the Commission grasped the opportunity to broaden its activities substantially. This opportunity arose because there was a widespread preoccupation that Europe was falling behind Japan and the U.S. in new technologies, something that was discussed under the lable of "Eurosklerosis" (Nussbaum 1984) in the early 1980s. Apart from continuing its activities to restructure the ailing steel industry, the EC Commission set out to conquer new policy fields, particulary in technology policy. In fact, the new thrust in this field was an important factor in the re-vitalisation of the EC that had appeared moribund to some observers in the early 1980s. The issue at stake today is to decide what sort of measures are worth continuing, which should be discontinued, and what might be promising new approaches.
What is the track record of European industrial policy?
Looking back at about ten years of a somewhat broader European industrial policy, two main conclusions emerge. Firstly, these years were marked by policy experimentation and learning. The Commission of the EU experimented with a number of different policy instruments, and it did not only learn from its own experiences but also tried to utilize scientific support for preparation and evaluation of projects (e.g. in the FAST programme). A highly differentiated and complicated pattern of political arenas emerged, involving other actors like the European parliament, national governments, firms and business associations, other pressure groups, and the scientific community. As ten years are a comparatively brief period of time to establish new policies, decision-making structures, and institutions from scratch, it is not at all surprising that the industrial policy has only been partially successful.
Secondly, it has become obvious that European industrial policy is no easy venture at all. Rather, policy makers constantly have to struggle with certain limitations and conflicting goals and interests. In fact, it may be much more difficult to explain the successes than the failures of European industrial policy, given that the extreme diversity of national political, economical, social, and cultural structures turns the formulation, let alone the implementation, of coherent policies into a rather adventurous endeavour. This is not made easier by the fact that it is not totally clear what is meant by industrial policy. A pragmatical solution is to define industrial policy as those measures that are explicitly meant to shape the industrial structure, i.e. to alter the outcome of the market lottery. There are at least five policy fields which fall into this definition, namely technology policy, regional policy, foreign trade policy, competition policy, and standardisation policy. (Buigues and Sapir 1993, 25 ff, and Nicolaides 1993, 12 ff)
There are two factors behind the mixed results of European industrial policy. Firstly, it is part of an ongoing learning process. It can hardly be expected to work perfectly from the onset, and there is certain evidence that the EU commission is learning from its past mistakes. Secondly, formulating a coherent industrial policy is an awfully difficult, if not outright impossible venture since there is a number of contradictions that are hard to resolve. There are five such contradictions:
There is no consensus in the EU that there should be industrial policy. The British conservative government has always been hostile to any kind of interventionist policy, be it on the national or the community level (except in the infamous banana trade regime case, but this is another story, and it is, strictly speaking, no industry issue). (Pedler 1994) The conservative government in Germany has always found it difficult to make up its mind, proclaiming free market ideology on Sundays and pursuing a pragmatic approach from Monday to Friday. However, schizophrenia can hardly serve as a solid base for coherent policy-making. Other governments, namely the French and the Italian, have always preferred a strong State presence in the economy, be it with State enterprises or via governmental guidance. (Buigues and Sapir 1993) Similarly, the other Mediterranean States have a long tradition of far-reaching State activities in their latecomer industrialisation effort.
The EU Commission reflects the different national styles. Although commissioners should be independent, they have shown a certain tendency to follow the respective national styles. (Ross 1993) As head of the Commission, Jacques Delors has advocated active industrial policies. Italian commissioner Pandolfi heavily pushed the HDTV project, whereas his British colleague strictly opposed it (although in this case not only for ideological reasons), and the Germans found it hard to define their position.
The diverging views have made it difficult to elaborate a coherent industrial strategy for the EU. As soon as somebody proposed such a strategy, a discussion started on whether or not there should be active industrial policy at all, rather than about its shape and general direction. An elementary consensus about the shape of industrial policy measures is emerging only slowly. Critics of industrial policy accept that there may be situations in which it makes sense to intervene whereas the supporters of industrial policy refrain from highly distorting programmes (especially if they discriminate in favour of large, firmly established firms). Also, the Fortress Europe-related ideas of focusing industrial policy on pure European firms appear to vanish. Apparently, this reflects the insight that the firms increasingly try to position themselves in all important markets, and enter into alliances with U.S. and Japanese firms since this is the only way of mastering leading-edge technology.
Diffusion-orientation vs. mission-orientation
European countries have different traditions in terms of industrial policy. Following the typology of Ergas (1987), France and the UK have preferred mission-oriented policies whereas Germany and smaller countries like Denmark have a tradition of diffusion-oriented industrial policy. The main feature of mission-oriented policies is that they focus on a small number of big national projects around which industries cluster and technologies emerge. Basically, it is a kind of big-push approach. Examples are large military programmes, e.g. missiles, the Concorde programme, or the French nuclear reactor programme.
Diffusion-oriented policies try to help modernizing a wide array of industries. They establish networks of technology research laboratories and dissemination agencies, not focusing exclusively on high technology but rather trying to upgrade all kinds of industries.
The two approaches are not necessarily mutually exclusive. Rather, each has its respective strengths. The mission- oriented approach is strong in stimulating new and emerging technologies and in fostering technological breakthroughs. The diffusion-oriented approach is strong in supporting incremental innovation within clearly defined trajectories. However, experience has shown that it is often unclear—even among key actors—which of the two approaches is behind certain policy initiatives. The addressees of the policies, namely private firms and research labs, often find it difficult to understand what the underlying rationale of a given policy is. Therefore, it is difficult for them to respond adequately.
It is also often complicated to exploit potential synergies between programmes if some of them follow the first and other the second approach. For instance, this has been the case in information technology where the linkages between the HDTV project (a typical case of a mission) and the ESPRIT programme (more diffusion-oriented) were weak.
Protection of sunset industries vs. stimulation of high-tech
One of the core dilemmas of industrial policy is that it is strongly influenced by pressure groups, and that declining industries are politically much firmer established than emerging industries. Industrial policy makers find themselves under constant pressure to implement policies that take the burden of adjustment off the shoulders of firms in declining industries. Since industries tend to be clustered, it is not only firms but rather regional policy clusters made up by politicians, bureaucrats, representatives of trade associations and of firms that exert pressure.
The track record of European industrial policy for declining industries has not been too impressive because policy makers have not been able to deal adequately with this dilemma. (Stehn 1993.) In the case of declining industries, they have by and large not succeeded in managing capacity reduction in a way that comes anywhere close to the Japanese experience. Most probably, this has been due to the fact that there simply was no easily acceptable solution, as industries in different member countries with vastly diverging levels of efficiency and modernity were involved. Typically, firms in the stronger member countries were more efficient and innovative than those in weaker countries, so that performance-related restructuring would have implied a further weakening of already weak countries, something that was politically not acceptable for them.
On the other hand, entrepreneurs in emerging industries feel little pressure to organize themselves and to look for political support. Industries and firms are growing rapidly and therefore feel no need to fight for their interests. There is no immediate logic for collective action under such circumstances. However, policy makers may strive to support such industries in order to defend common interests. But they face the problem that it is difficult to establish adequate communication links with new firms exactly because they have not (yet) organized themselves. This is a reason why European industrial policy for emerging industries has more often than not focused on long-established, large corporations that entered into new fields since it was much easier to establish a communication network with them; and adequate information about the problems and desires of a target group is a core prerequisite for successful industrial policy.
International competitiveness vs. cohesion
Industrial policy decision-making in the EU is further complicated by the fact that there are two conflicting aims, namely strenghthening the international competitiveness of the EU industries and reducing the imbalances between the regions of the EU, i.e. fostering cohesion. The first aim implies strenghthening the already strong, the second supporting the weak. Whatever priority the Commission sets is bound to face resistance from either the strong or the weak member States.
Commission programmes have tried to reconcile these two aims, yet this has not been too successful. For instance, in certain technology programmes only those applicants qualified for EU support that formed a consortium including firms or laboratories from weaker member States. Experience shows that this tended to destabilize consortia, or to reduce their overall performance, or that weaker consortium members were simply marginalized during the operational phase. (Starbatty and Vetterlein 1989) One of the reasons for launching the EUREKA initiative (which is technically outside the domain of the EU) was that the stronger member States wanted to have a programme where the competitiveness-orientation was not hampered by cohesion-orientation.
In other cases, attempts of reconciliation led to political deal-striking where industrial policy was linked to completely different issues, e.g. weaker member States demanding certain concessions in agricultural policy in change for their support of competitiveness-oriented industrial policy initiatives.
Another problem is that the goal of cohesion policy is not entirely clear: Does it try to homogenize the member States, that is help the weaker member countries to replicate the development experience of the more advanced countries, or does it rather aim at stimulating the weaker countries to find their own way to prosperity? The latter view—albeit being vague—may offer an opportunity for a new start. This is all the more necessary since traditional regional policy that largely operated with financial instruments had, at best, a limited impact whereever it was practiced, and therefore does not offer any perspective of overcoming regional disparities. At the same time, as Fisch (1993) has pointed out, there is good reason to believe that a certain degree of heterogeneity raises the level of welfare in the community as a whole, for instance because one of the outcomes is complementary trade, which is less prone to cause conflicts than substitutive trade. Therefore, regional policy should help backward regions in developing a specialisation profile. If they want to develop competitive advantages, their firms must be faster, better and cheaper than other suppliers in certain industries, or in service sectors. In order to master these challenges, local policy makers need, first of all, imagination and creativity. Without that, even loads of funds will only have a minor effect.
Centralisation vs. localisation
It is largely unclear how tasks should best be distributed between the Commission of the EC, the national governments, and regional/local governments. The magic term "subsidiarity" has been introduced to resolve this puzzle, but unfortunately it does not solve anything because it has not yet been defined. (Sinn 1993) It is not even certain that a clear-cut definition is possible at all. Although the basic meaning may be clear, namely as little centralisation as necessary, anybody is free to define which tasks require centralisation and which do not according to his or her preferences and interests-and some observers have interpreted the impetus of the White Paper on Growth, Competitiveness, and Employment as an effort to shift even more reponsibilities to the Commission (H.König, Ein Weißbuch für den Interventionismus, FAZ, 29.7.94.)
In practical terms, the Commission has shown a tendency in the past to accumulate as many competences as possible. This is particularly true in fields like technology policy where the then-EC had practically no mandate at all until 1986 when the Single European Act was passed; still, it started to formulate technology initiatives earlier. At the same time, however, it must be mentioned that the Commission also tries to stimulate learning processes between European regions so that the less successful can learn from the positive experiences of the more successful. Moreover, it must be pointed out that critics of Eurocracy and centralization often miss the point—the Commission acts more as a coordinator and stimulator than like a super-government of the traditional, hierarchical style.
Scenarios for industrial policy
Reasoning on industrial policy can be embedded into different scenarios. I will describe three of them: the attempt to avoid industrial policy at all, traditional hierarchical and lobby-driven industrial policy, and a sort of new industrial policy based on new modes of governance.
Scenario 1: No industrial policy
There are governments in Europe that despise industrial policy, based on the conviction that it is a distorting and hence costly exercise. This applies particularly to the current governments in the UK and Germany. Yet, there can be little doubt that the German government is pursuing industrial policies, in particular for sectors in crisis. Even in the new Länder, where the federal government deliberately refrained from formulating an industrial policy in the beginning, there are now policy measures that aim at certain branches and even on individual firms. (Kern 1994.)
Behind this is, basically, the fact that politicians, being politicians, are under constant pressure to react to certain problems, bottlenecks, and crises that emerge within society. More often than not, industries in crisis are not scattered across a country but rather clustered in a certain region (or a limited number of regions). Therefore, a crisis that affects a given industrial branch often affects one or some regions and may indeed threaten the very basis of prosperity in this region. In a centralized economy, local politicians, firms, trade unions, and possible other actors will mobilize to exert pressure on central government to cushion the adjustment process. In a federal economy, the representatives of the constituency in the federal parliament will immediately feel the pressure and will most probably see themselves as a defender of the interests of their home region. In both cases, government will hardly be in a position to refrain from interventions on the grounds that this would be suboptimal in macroeconomic terms. It probably will not declare its support measures as industrial policy but rather as structural policy, or labour market policy, or it may indeed camouflage them as infrastructural or public purchasing policy. In any case, however, it is going to implement measures that come very close to genuine industrial policy.
It is obvious that, at least under democratic conditions, it is virtually impossible to refrain completely from implementing industrial policy. In fact, the neo-liberal stand against industrial policy has disastrous effects for two reasons. First, it leads to an industrial policy that is mainly concerned with ailing industries. This is exactly the case, for instance, in Germany where old industries (coal, steel, shipyards) receive much more support than emerging industries with a high growth potential and positive externalities. This is due to the fact that old industries are much better in organizing pressure for political support—their common interest and therefore the logic of collection action is obvious (and much more obvious than for individual entrepreneurs in new industries who are more prone to see individual achievement as the basis of economic success), they have well-established political contacts, and they have a strong bargaing position due to the large number of jobs that may be threatened.
Second, a reactive industrial policy that is formulated as a response to an immediate crisis will usually operate on the basis of ad-hoc-measures which will often be measures whose ineffectiveness is proven but which still appeal to politicians because they appear plausible. Ad-hoc industrial policy by definition is not open to systematic evaluation since it is supposed to be a short-term, immediate-impact exercise and therefore does not allow for continuous learning.
Having no industrial policy at all is no realistic option. The real alternatives are a lobby-driven, reactive industrial policy, and a network-driven, anticipative industrial policy.
Scenario 2: A lobby-driven, reactive industrial policy
This is exactly the kind of industrial policy that has provoked all kinds of criticisms due to its intransparency, unpredictability, and arbitrariness. It incurs high costs, and since its aims often are unclear, there is no mechanism of phasing out State support built into programmes; in fact, there are cases like German coal subsidies where programmes run for a generation. Add to this the appalling appearance of huge sums of money going to ailing and/or big firms, and it becomes obvious why some economists and politicians tend to get emotional about the issue.
Indeed, the sort of industrial policy that saves firms for the sake of keeping jobs, that allocates high subsidies for technological research to firms which are not exactly poor (in Germany the saying goes that Siemens is the only bank with an attached electronics production), and which channels funds to regions that are already economically strong is difficult to justify in economic terms. Its existence is rather due to political factors, but, alas, it is not even very likely to achieve the proclaimed political goals.
For instance, it is difficult to measure the outcomes of industrial policy. Therefore, just like in technology policy where input variables like patents or R&D expenditure are taken as proxies for the probable output, in industrial policy input variables like the absolute amount of State subsidies are an obvious candidate for the role of the indicator of success, at least when it comes to demonstrating achievements to those who demand that something is being done. (Peterson 1993, p. 207.) A stupid but high subsidy will hardly go unnoticed—on the side of the critics or non-beneficiaries for exactly the reason of being stupid, on the side of the beneficiaries on the grounds that a huge subsidy simply must have some sort of positive impact. On the other hand, smart interventions that support sustained competitiveness but do not have an immediate impact have little value for a politician who has to demonstrate to his constituency that immediate action is being taken.
And yet it is very difficult to get rid of this kind of industrial policy. For politicians who want to be re-elected it is still better than no policy at all; firms will always look for such sources of income; and those who are to suffer from these policies are not directly affected but rather suffer from the negative impact on economic growth, and will thus not organize their countervailing interests. Isolated reforms that aim at procedures, that is try to to establish clearer rules for policy measures, will not improve the situation because they do not alleviate the pressure under which politicians act. Demanding an end to all industrial policy is no serious option, either, for the reasons outlined above. The remaining option is a fairly fundamental reformulation of the concept of industrial policy; this can be based on successful examples in EU policy, on the national level, and in non-European countries.
Scenario 3: A network-driven, anticipative industrial policy
The discussion on whether or not there should be industrial policy at all often starts from the assumption that industrial policy implies a central government trying to steer economic processes. In fact, this idea has flourished in the heighdays of Keynesianism and social engineering in the 1970s. Critics of industrial policy, however, tend to ignore the fact that this is not the only way to govern modern, and this means highly heterogeneous and differentiated, societies. In fact, old-style hierarchical governance is dead and gone, at least by and large. Instead, two different types of governance have emerged. The first, a laissez-faire approach, has been prominent in the UK. It has been successful in abolishing obstacles to industrial modernisation, but it has largely failed (and actually not really intended) to create an environment that supports industrial competitiveness. Leaving each and everything to anonymous market forces has proved to be no successful recipe to revitalize industrial dynamism.
The second type of governance appears to be more promising. So far, it lacks a clear label; one attempt has been heterarchic, another network governance. This type has been particularly prominent on the local level and inside regions in some European countries where the State has not guided the action but served as a facilitator and moderator. But is has also been applied on the national level. The characteristic feature of this type of governance is the emergence of policy networks that are made up by civil servants, representatives of private business, trade unions, the scientific community, and other societal groups. The core elements of policy networks are elements like fair exchange, reciprocity, or a just distribution of the costs and benefits of a joint decision or a given problem solution. These are demanding conditions, but they are also rewarding since they result in superior industrial environments which sustain a high level of welfare.
This kind of approach has at the same time clear advantages and disadvantages. One of the key advantages is its potential to solve the key shortcomings of laissez-faire and hierarchic governance. Laissez-faire suffers from the fact that market failures are not corrected; hierarchic governance suffers from the fact the government often lacks the necessary information to formulate appropriate policy measures. Provided that the key actors, that is representatives of firms, industry associations, trade unions, and other non-governmental organisations, as well as State representatives, are involved, policy measures being defined within policy networks can overcome these problems. Target groups are more likely to understand and to accept policy measures if they have been involved in their formulation. Policy measures formulated in such a way will often also aim more clearly at key bottlenecks.
The obvious disadvantage of policy networks is their potential to fail. Network failures occur, for instance, when the network is too close, that is if competing interests are not present; in this case the probability that actors will strive for rent-seeking and/or externalize the costs of certain policy measures is extremely high. Policy networks should thus be embedded into hierarchical structures—there should be a higher level of government that supervises the composition and procedural rules (but not necessarily the policy formulation itself) of policy networks. This is particularly important in the case of the EU where groups that have a well-established presence in national-level policy networks find it difficult to enter into EU-level networks; the trade unions give an example. For EU technocrats involved in industrial policy making, the temptation to build networks that only include firms, academics, and certain business associations (namely those where a limited number of mostly large firms is capable of formulating a concise position) is strong—firms and academics, and in some cases business associations, can provide them with the know-how they need to draft policy proposals, or indeed the ready formulated proposals themselves, (Middlemas 1994, p. 5.) whereas peak associations of trade unions, and also certain other business associations, have little to offer in this respect. This reflects the fact that the latter are typically made up by members from different countries with different backgrounds, interests, and policy prescriptions. Accordingly, even if they may have the technical competence to prepare a concise draft proposal (which they, in fact, often lack), it is quite complicated for them to organize internally a process to find a common position that has the necessary substance, i.e. goes beyond a minimum consensus. The holds not only true for trade unions but also consumer organizations and other associations. Thus, the more difficult it is to organize interests on the European level, the less probable it is that these interests are going to be involved in the policy-making process.
The conclusion is that network-like structures need not be newly constructed on the level of the EU; in fact, they have been in place for quite some time, being the arena where Commission officials, national civil servants, representatives of firms and business associations, scientists, and others generate ideas and formulate policies. (Schendelen 1994.) However, they lack so far a minimum amount of transparency, procedural rules, and control of composition. At best it is possible to describe patterns that govern certain sub-arenas at some point in time. The actors involved, and the coalitions between them, are constantly changing. It is sound to assume that the transaction costs for all parties involved are high, and higher than necessary, since the level of turbulence, complicatedness, and unpredictability is high. Moreover, the fact that the policy formulation process in these networks lacks any sort of supervision lies at the heart of the democratic deficit of the EU. The solution, however, should not be the dismantling of these networks; there is no more appropriate arrangement available that might substitute them. Rather, policy makers who want to increase the effectiveness of EU policies and at the same time tackle the democratic deficit should try to follow the principle of context governance, that is establish elementary rules that guide the composition of networks and the interaction inside and between them. Moreover, the EC Commission has played an active role in the formation of European associations, e.g. the European Information Technology Industry Roundtable, an association of the leading European electronics firms, (Grande 1994, p. 14.) or the European Roundtable of Industrialists, a group of about forty large European firms from different industrial sectors. (Green 1994, p. 22.) There is no a priori reason why the EU shouldn't encourage the emergence of European peak associations of other important actors, e.g. trade unions, consumers, or environmentalists.
Elements of a European industrial policy
Given the relatively brief period since its inception on a broad scale, the track record of European industrial policy is not too bad. Given the complicatedness of the requirements the Commission has to fulfill, and the divergence in national styles and conceptions, this is all the more surprising. It would therefore not only be unrealistic but also foolish to demand that the EU refrained from formulating and implementing industrial policies. Rather, policy makers in the member countries should pay closer attention to how the formulation process is actually taking place; and they might indeed decide to demand changes in procedural rules in order to secure more transparency and accountability.
It is also worthwhile to take the principle of subsidiarity more serious in this context. Currently the Commission is engaged in various programmes that aim at individual firms, including small and medium firms. In fact, the latter sort of programme has come about largely as a response to those who criticized the Commission for giving too much money to large firms and thus neglecting small and medium firms. However, formulating such policies on the Union level cannot be the best solution, given the diversity of firms' structures and problems. Cutting the subsidies for large firms would have been the better approach. Indeed, the costs of EU programmes for small and medium firms are probably high—on the level of the firm in terms of transaction costs since it takes a lot of effort to find out about EU programmes and to apply for EU funds, and in terms of social opportunity costs since many firms refrain from applying for support since they fear too much red tape; it is thus not at all clear (and, in fact, quite doubtful) that the funds flow to those firms who most deserve them.
Again, this is not a plea against industrial policy but rather for a more differentiated and systematic approach to it. Ideally, industrial policy should consist of three elements:
State intervention in cases of market failure, for instance if firms underinvest in R&D because they fear that they will not be able to appropriate the results of their effort. Yet it is important to differentiate between different types of R&D. The case for State support is strongest in early phases of radical technical change, that is when technological trajectories are not yet defined and the critical mass in terms of investment in R&D and production facilities does not appear. (Meyer-Stamer 1994) The EU has a significant role to play here since in fields like new telecommunication technologies trajectories must be defined on a supra-national level; European national markets are too small to stimulate the emergence of world-class competitors in such fields.
Recent proposals for European industrial policy in this field take up the classic Keynesian idea of digging holes, the major difference being that it shall not be for the sake of having holes but rather to lay more telephone cables. This is not necessarily a bad idea—it is going to create jobs, and there may be demand for more telecommunication capacity. However, when it comes to describing the potential demand for this capacity, documents tend to get hazy, mentioning phenomena like the 'transition to the information society' as a way of justifying more cables. It may appear that European policy makers have not entirely learnt the lesson of the failed HDTV project (which totally neglected the demand side). In fact, there are good reasons to assume that there is a certain amount of potential demand for additional telecommunication capacity; however, it is well possible that supply and demand are well balanced given the prevailing price structure. Laying new cables will therefore only stimulate higher demand if temporary oversupply pushes down prices. This will hardly happen under the conditions of telecom monopolies. Therefore, industrial policy in this field must necessarily address the regulatory framework (and must, indeed, induce changes in the regulation pretty fast). The Commission is trying to push this process forward, (Bangemann-post-Corfu, I.3) but it is facing the resistance of several member country governments.
Even if this is going to change in the near future, another problem arises. There is reason to believe that liberalisation will mainly benefit corporate customers, whereas the position of consumers will deteriorate. So far, it is not clear whether this is something the Commission desires, or is willing to tolerate. A call for more creativity in designing future communication infrastructures might not be a bad idea. Issues like guaranteed access to and citizen participation in the development of new information services are largely missing in the European discussion, particularly compared to current discussions in the U.S. True, the Commission plans to launch studies on the social impact of new communication technologies; but this is not articulated with the planning process where the lack of consumer involvement is obvious. Things may go badly wrong over the next years—and thanks to the social impact studies, we will ex post know precisely why they have gone wrong.
There is yet anther problem associated with the new emphasis on telecommunications. Even though the Commission had tried to correct this bias, European industrial policy has mainly addressed big firms for various reasons. The bias may be reinforced with the current approach to telecommunications that aims at big investments. This implies neglecting the key shortcoming of European industry, especially compared to the U.S., namely the insufficient generation rate of new firms. The absence of the European equivalents of the Apples, Suns, Silicon Graphics, or Oracles is mainly due to deficiencies in the institutional settings in the member countries—the absence of 'patient capital' in the UK, and the preference for big or well-established firms and a very conservative concept of collateral in countries like Germany. (Sharp and Pavitt 1993, p. 142.) There is little the Commission can do about this in the short run. In particular, one should not expect EU-financed venture capital funds (Rothwell and Dodgson 1992, p. 229.) to solve this problem since they have to be channeled through the same risk-averse financial institutions that have contributed to the limited success of similar national programmes. Yet, longer-term policies should specifically try to address these bottlenecks, in particular by addressing structural impediments that result from the conservative behaviour of old, large firms and banks.
Another obvious case of market failure is environmental damage. When it came to environmental regulation, however, the Commission met fierce resistance from member countries. It was only in the 1993 White Paper on Growth, Competitiveness, and Employment that a shift to an environmentally sound industrialisation model was proposed; however, the respective part of the White Paper was not too well articulated with the other chapters. In trying to define this concept more clearly, the EU could kill several birds with one stone—not only improving living conditions, but also increasing pressure on European firms to increase their efficiency, especially their eco-efficiency, and creating new markets for things like microelectronic devices to improve the environmental soundness of production processes and consumer products. In fact, in a situation where no national government wants to be a first mover when it comes to creative environmental policy formulation a space opens for the Commission by defining common, community-wide rules that do at least create a level playing field within the EU.
Locational policy to strenghthen the specialized factors that are the basis of competitive advantages (following the terminology of Porter). ( Porter 1990) In this field, the EU has to play a complementary role to local and regional initiatives. It is on the local level that specialized factors are being created, namely in terms of supporting institutions in technology, training, or financing. (Schmitz and Musyck 1993) This is not only true for strong regions, but also for declining regions that strive to achieve a turnaround. (Price, Morgan, and Cooke 1994) The EU can stimulate local policy initiatives by encouraging exchange of experiences between European regions or by introducing incentives for local policy formulation.
There are quite a number of possible actors in the field of industrial policy—not only the Commission but also actors on the national and different levels of local government. Experiences in the 1980s have shown that the regional and local levels have considerable potentials. Regional and local governments have played an active role in managing structural change and supporting enterprise competitiveness in areas like Rhône-Alpes in France, in Baden- Württemberg and Nordrhein-Westfalen in Germany, in Wales, and in various industrial districts in Northern Italy. (Cooke 1992, Schmitz and Musyck 1993) There is little reason to believe that the EU could be better in doing this; this is a further argument against firm-related programmes on the EU level. What the Commission should rather do is to stimulate local and regional governments to formulate their own industrial locational policies, to support this by mobilizing expertise and (to a limited extent) funds, and to stimulate learning between regions. Things like technology diffusion or specialised service provision can probably best be managed by local governments, especially in those cases where specialized industrial clusters are able to articulate their specific demands for supporting institutions. In fact, the national governments will have the biggest difficulties in redefining their role. (Grande 1993) They should refrain from directly intervening on the local level and rather stimulate autonomous policy formulation on the regional and local levels, too. This is not to be confused with traditional regional policy where the centre tried to create dynamism in lagging regions. This approach has largely failed.
Management of structural adjustment, i.e. State intervention that stimulates and moderates capacity reduction in sunset industries, thus inhibiting a chaotic, overly destructive adjustment process and at the same time avoiding too high levels of protection for moribund firms. For the reasons mentioned above, this is a particularly difficult venture. Since workers and shareholders in rising and vigorous industries have to pay for the subsidies to declining industries, anybody who wants to rationalize the management of adjustment must try to build an alliance with them in order to get the necessary political support. Better information on the costs of subsidies may help to do this. The EU could play an important role in increasing the level of transparency about such policies, i.e. by collecting and disseminating information on local, national and supra-national subsidies and thereby raising the awareness for the costs of traditional adjustment policies. More up-to-date information would enhance transparency and shift the balance between the disguising activities of the parties involved and the uncover activities of those who have to pay the bill.
Recent policy changes like the conditional approval of national subsidies for airlines in several member countries, with the conditionality being quite strict, indicate that the Commission has learnt its lessons. It remains to be seen whether the Commission can maintain its policy that the most recent round of approvals was the last one. If it does, this would be a major achievement—and one that may help improve the often unjustly bad perception of European industrial policy.
ABBING, M.R., and SCHAKENRAAD, J. (1990). Joint R&D Activities of Firms in European Cost-Sharing Programmes. Maastricht: MERIT.
BUIGUES, P., and SAPIR, A. (1993). Community Industrial Policies. In P.Nicolaides (Ed.), Industrial Policy in the European Community: A Necessary Response to Economic Integration? (pp. 21-37). Dordrecht etc.: Nijhoff.
COOKE, P. (1992). Regional Innovation Systems: Competitive Regulation in the New Europe. Geoforum, Vol. 23, No. 3,pp.365-382.
COWLES, M.G. (1994). Organizing Industrial Coalitions. Paper presented at the Sussex European Institute Seminar 'Industrial Networks in the EC' 13-15 October.
DUNNING, J.H. (1993). Governments and Multinational Enterprises: From Confrontation to Co-operation? In L.Eden, E.H.Potter (Ed.), Multinationals in the Global Political Economy (pp. 59-83). New York: St.Martins Press.
ERGAS, H. (1987). Does technology policy matter? In B.R.Guile, H.Brooks (Eds.), Technology and global industry. Companies and nations in the world economy (pp. 191-245). Washington, D.C.: National Academy Press.
FISCH, G. (1993). Außenhandels- und entwicklungstheoretische Aspekte der Kohäsionspolitik der EU. Konjunkturpolitik, Jg. 39, Nr. 6, S.349-375.
GRANDE, E. (1993). Die neue Architektur des Staates. Aufbau und Transformation nationalstaatlicher Handlungskapazität -untersucht am Beispiel der Forschungs- und Technologiepolitik. In R.Czada, M.G.Schmidt (Eds.), Verhandlungsdemokratie, Interessenvermittlung, Regierbarkeit (pp. 51-71). Opladen: Westdeutscher Verlag.
GRANDE, E. (1994). Die Technologiepolitik der EU: Fragestellungen und Forschungsperspektiven für die Politikwissenschaft. Beitrag für den Arbeitskreis 'Politik und Technik' des 19.DVPW-Kongresses Potsdam, 25.-28.8.
KERN, H. (1994). Intelligente Regulierung. Gewerkschaftliche Beiträge in Ost und West zur Erneuerung des deutschen Produktionsmodells. Soziale Welt, Jg. 45, Nr. 1, S.33-59.
LEHMANN, J.P. (1992). France, Japan, Europe, and industrial competition: the automotive case. International Affairs, Vol. 68, No. 1.
McALEAVEY, P.C. (1994). Who is the operative subject of EC Regional Policy? The cases of Scotland and North Rhine-Westphalia. In U. Bullmann (Ed.), Die Politik der dritten Ebene: Regionen im Europa der Union (pp. 79- 90). Baden-Baden: Nomos.
MEYER-STAMER, J. (1994): Steuerungsprobleme von Medien- und Technikentwicklung. In: O. Jarren (Hrsg.), Medienwandel - Gesellschaftswandel? 10 Jahre dualer Rundfunk in Deutschland. Eine Bilanz. Berlin: Vistas. S.131-46.
MIDDLEMAS, K. (1994). Informal Politics: Power without Sovereignty. Paper presented at the Sussex European Institute Seminar 'Industrial Networks in the EC' 13-15 October.
MYTELKA, L.K. (1993). Strengthening the Relevance of European Science and Technology Programmes to Industrial Competitiveness: The Case of ESPRIT. In M. Humbert (Ed.), The impact of globalization on Europe's firms and industries (pp. 56-63). London: Pinter.
NICOLAIDES, P. (1993). Industrial Policy: The Problem of Reconciling Definitions, Intentions and Effects. In P.Nicolaides (Ed.), Industrial Policy in the European Community: A Necessary Response to Economic Integration? (pp. 1-17). Dordrecht etc.: Nijhoff.
NUSSBAUM, B. (1984). Das Ende unserer Zukunft. München: Kindler.
PEDLER, R.H. (1994). The Fruit Companies and the Banana Trade Regime (BTR). In R.H. Pedler, M. van Schendelen (Ed.), Lobbying the European Union (pp. 67-92). Hants: Dartmouth.
PETERSON, J. (1993). High Technology and the Competition State. An analysis of the Eureka initiative. London, New York: Routledge.
PORTER, M.E. (1990). The Competitive Advantage of Nations. New York: The Free Press.
PRICE, A., MORGAN, K., and COOKE, P. (1994). The Welsh Renaissance: Inward Investment and Industrial Innovation. Cardiff: CASS.
ROSS, G. (1993). Sidling Into Industrial Policy: Inside the European Commission. French Politics & Society, Vol. 11, No. 1, pp.20-43.
ROTHWELL, R., and DODGSON, M. (1992). European technology policy evolution: convergence towards SMEs and regional technology transfer. Technovation, Vol. 12, No. 4, pp.223-38.
SCHARPF, F.W. (1992). Wege aus der Sackgasse. Europa: Zentralisierung und Dezentralisierung. WZB-Mitteilungen, Nr. 56, S.24-28.
SCHMITZ, H., and MUSYCK, B. (1993). Industrial Districts in Europe: Policy Lessons for Developing Countries?. Brighton: Institute of Development Studies.
SHARP, M., and PAVITT, K. (1993). Technology Policy in the 1990s: Old Trends and New Realities. Journal of Common Market Studies, Vol. 31, No. 2, pp.129-51.
SINN, H.W. (1993). How Much Europe? Subsidiarity, Centralization and Fiscal Competition. London: CEPR.
SITTE, R., and ZIEGLER, A. (1994). Die EU-Strukturfonds nach der Reform. WSI-Mitteilungen, Jg. 47, Nr. 4, S.214- 22.
STARBATTY, J., and VETTERLEIN, U. (1989). Spitzentechnologie oder innere Kohäsion. Ein technologiepolitischer Zielkonflikt in der Europäischen Gemeinschaft. Europa-Archiv, Nr. 5, S.145-54.
STEHN, J. (1993). Wettbewerbsverfälschungen im Binnenmarkt: Ungelöste Probleme nach Maastricht. Die Weltwirtschaft, H.1, S.43-60.
STURM, R. (1992). Konkurrenz oder Synergie? Nationale und europäische Industriepolitik. In M.Kreile (Ed.), Die Integration Europas (PVS. Sonderheft 23) (pp. 234-53). Opladen: Westdeutscher Verlag.
STURM, R. (1992). Regionalisierung der Industriepolitik? Politik und Zeitgeschichte, B 10-11.
THOMAS, K.P. (1993). EU Regulation of State Aid to Industry: Lessons for North America. St. Louis (mimeo).
VAN SCHENDELEN, M.P.C.M. (1994). Studying EU Public Affairs Cases: Does it Matter? In R.H.Pedler & M. van Schendelen (Eds.), Lobbying the European Union (pp. 3-20). Hants: Dartmouth.
WERLE, R. (1993). Politische Techniksteuerung durch europäische Standardisierung? In H. Kubicek, P. Seeger (Ed.), Perspektive Techniksteuerung (pp. 129-142). Berlin: Ed. Sigma.