Why increasing transparency in the European Union will not make lobbyists behave any better than they already do Paper prepared for the EUSA Ninth Biennial International Conference Austin, March 31-April 2, 2005 Daniel Naurin Dep. of Political Science, Goteborg University and The Swedish Institute of Foreign Affairs, Stockholm daniel.naurin@pol.gu.se Abstract According to deliberative democratic theorists transparency has a civilizing effect on political behaviour, forcing actors to argue public-regardingly instead of engaging in self-interested bargaining. Negotiation theorists, on the other hand, warn that transparency may damage effective problem-solving and lead to harder group polarization. This paper scrutinizes this debate, points at the different assumptions underlying the competing hypothesis and subsequently tests these in a comparative study of business lobbyists acting in institutional settings characterised by varying degrees of transparency in Sweden and the European Union. The results support the negotiation theory hypothesis, but it is also seen that a deliberative perspective is necessary to understand the behaviour of lobbyists backstage and that standard two-level games often will be inadequate for understanding transparency effects. What is the effect of opening up secluded processes of political decision-making to a public audience, by means of transparency reforms? A common view in the debate on the democratic status and legitimacy of international organisations, the European Union in particular, is that increased transparency is a must. Transparency is believed to strengthen public confidence in political institutions and increase the possibilities of citizens of holding decision-makers accountable for their actions. Furthermore, which is especially important for institutions that lack normal democratic procedures, transparency is said to promote "output-oriented" legitimacy . Transparency will prevent "wrong-doing" , make decision-makers more "responsive" to the public and "secure the adoption of more impartial decisions" . Since institutional structures for "input-oriented" legitimacy (i.e. democratic political institutions) are weak at the international level, such output-oriented legitimacy may be a welcome compensation; a "substitute legitimation" in Adrienne Héritier's words. The drive towards increasing transparency is not applauded everywhere, however. Especially those actors who are potentially exposed to publicity, such as governmental negotiators themselves, tend to be sceptical. A certain degree of secrecy is necessary in order to reach agreements, they argue. Transparency puts the effectiveness of international negotiations at risk. For instance, the famous culture of compromise in the European Union Council of Ministers may be damaged if radical transparency reforms are implemented there. This article investigates the debate over the effects of transparency and publicity on elite political behaviour. The advocates of transparency find strong support in deliberative democratic theory, where publicity is seen as "one of the purifying elements of politics". Sceptics, on the other hand, may lean towards negotiation theory, where it is argued that "a good case can be made for changing Woodrow Wilson's slogan 'open covenants openly arrived at' to 'open covenants privately arrived at'". In order for negotiating parties to reach a "wise outcome", according to negotiation theorists, "it is useful to establish private and confidential means of communicating with the other side." . The nature of this theoretical debate is scrutinized in the first part of the article. It turns out that the two theoretical perspectives are only partly in disagreement about the effects of transparency. The difference in views with respect to the normative attractiveness of transparency reforms stems in part from the fact that negotiation theorists' main focus is on effects of transparency on positions, while deliberative theorists rather emphasise effects on arguments. But there is also a real disagreement about what the effect of transparency is most likely to be, which concerns the mode of communication and the types of justification. While deliberative theorists hypothesise that transparency and publicity will promote a shift away from self-interested bargaining towards arguing with public-regarding justifications, negotiation theorists would predict that the degree of self-interested posturing will increase in public. This disagreement, it is showed, is based on different assumptions about the characteristics of the public audience. The two different hypotheses are subsequently tested in a comparative study of business lobbyists acting in institutional settings characterised by varying degrees of transparency and publicity. The data consist of interviews with the best, according to reputation, public affairs consultants in Brussels and in Stockholm, focusing on what is appropriate and successful behaviour in different contexts. Sweden and the EU can be described as two extreme points on the transparency scale. The data also includes a unique sample of confidential lobbying letters from business organisations to the European Commission, which is compared to public communications from the same senders and publicly available letters from similar senders in Stockholm. The results clearly support the negotiation theory hypothesis. However, it is also seen that while deliberative theorists idealizes the public frontstage a deliberative perspective is necessary to understand the behaviour of lobbyists backstage. It is found that the opaque lobby corridors of Brussels are not arenas for striking deals, but for arguing about the best public policy. Furthermore, the standard two-level game perspective within negotiation theory is inadequate for understanding transparency effects, given a public sphere with at least some deliberative qualities. Deliberative democratic theory's case in favour of transparency The idea that public exposure has a civilizing effect on political action has a long history in political philosophy. Jeremy Bentham defended publicity on utilitarian grounds, arguing that it motivates public officials to do their duty. John Stuart Mill supported the open ballot on the basis that it would put a check on "the selfishness, or the selfish partialities, of the voter". In the last two decades the suggestion that publicity has the power to censor selfish and immoral behaviour and promote considerations of the common good has had something of a comeback with the "deliberative turn" in democratic theory. A common assumption among deliberative democratic theorists is that arguing, rather than bargaining or voting, is at the heart of democracy. Drawing on Jurgen Habermas's notion of communicative rationality it is assumed that politics should be—and can be—more about giving good reasons than forcing or striking deals. Assuming accountability for one's political actions by justifying them to one's fellow citizens is a question of treating those fellow citizens with respect. "When properly conducted", in Joshua Cohen's words, "democratic politics involves public deliberation focused on the common good". Deliberative theorists often position themselves in contrast to economic conceptions of democracy, where politics is "understood mainly in terms of conflict of competing interests—and thus more in terms of bargaining than of public reason". The role of transparency in deliberative democratic theory can be described as an institutional instrument that works to liberate politics from illegitimate 'market' behaviour. When political actors are exposed to a public audience they have to face two social norms, the deliberative literature (implicitly) assumes, which set the standard for behaviour at 'the forum' (the political sphere of society). First, there is a force-of-the-better-argument-norm, which regulates the mode of communication. The appropriate mode of communication in politics, this norm says, is arguing (seeking transformation of preferences via rational arguments) rather than bargaining (seeking aggregation of preferences via the exchange of threats and promises). Second, there is also a non-selfishness norm, which states that the appropriate types of justification for public policy are public-regarding rather than self-regarding. The combined effect of these two norms is to incur social costs on those actors who do not shift from self-interested bargaining to public-spirited arguing, when previously closed doors are opened for a public audience. Initially, this shift may amount to no more than mere rhetorical hypocrisy. The creativity of political actors in finding public-interested justifications for their self-interested positions is often impressive. However, deliberative theorists continue, rhetoric is not 'just rhetoric'. The civilizing effect hits not only the arguments, but also the actual decisions taken, because the public can only be fooled up to a certain extent. If one cannot demonstrate a reasonable connection between the public-regarding argument one is using and the position one has taken, there is a risk that one is revealed both as self-interested and as a hypocrite. Therefore, the policy proposals defended will have to be modified as a result of public exposure. Actors are forced not only to 'talk the talk', but also to 'walk the walk'. Transparency, thus, from a deliberative point of view, is an important component of the institutionalist program of designing "deliberative processes that favour broader over narrower interests [and] puts a premium on moral deliberation rather than power politics and bargaining". Negotiation theory's case against transparency The main problem with transparency and publicity, according to negotiation theorists, is that it impedes effective problem-solving capacity of political institutions. This is so primarily because actors become less willing to reveal private information and to change their positions once they have stated them publicly. Positive-sum "integrative bargaining"—creatively searching for compromises which all around the negotiating table may be satisfied with—negotiation theorists argue, requires that actors are prepared to reveal private information, including 'base motives'. [The parties] will not engage in problem-solving behavior unless the activity is relatively safe. Both Party and Opponent need to be assured that if they freely and openly acknowledge their problems, if they willingly explore any solution proposed, and if they candidly discuss their own preferences, this information will not somehow be used against them. […] The use of transcripts or a stenographer may inhibit exploratory and tentative discussions. Large galleries and disclosure to outside persons have the same effect. If actors do not feel safe about speaking their minds, brainstorming, throwing different options on the table, creativity in the problem-solving process is dampened. Furthermore, in the negotiation theory literature going public with a position implies giving it a higher "degree of finality". It is similar to saying 'this is our last offer, take it or leave it'. Positions which have been stated publicly become less flexible, meaning that the risk for negotiation breakdown increases. Making commitments with exactly the right degree of flexibility is part of the art of negotiation and being exposed to involuntary publicity makes this delicate task more difficult. There is, however, a possible counter strategy available to negotiators who are exposed to publicity, which may soften the hard-commitment effect. If the parties are exposed to unwanted publicity they could respond by lowering the "degree of specificity" of their positions. In that case they would try to counter the higher degree of finality of their statements, and the subsequent increasing inflexibility and risk for negotiation breakdown, by making them more vague. The less specific the public position, the less the risk of the negotiations getting stuck and the participants to losing face afterwards. The strategy of lowering the degree of specificity in order to safeguard flexibility, however, assumes that publicity is just temporary. The public frontstage, in that case, is not a decision-making arena, but an arena in which actors seek to enhance their status and powers, internally and externally, before the closed-door bargaining—where the real decisions are made—starts (or restarts). What is the controversy really about? A quick glimpse at this debate gives the impression that deliberative theorists and negotiation theorists are in full opposition with each other with respect to the effect of transparency reforms. This is partly an illusion. Those deliberative theorists who have looked at the issue more closely acknowledge that the process of reaching agreement through discussion tends to slow down in public as actors become more wary of what they say. For Jon Elster, who has developed the idea of publicity's civilizing effect most thoroughly within the deliberative literature, the negative effects on the quality of the discussions in terms of rational discourse amounts to a dilemma: "With total secrecy, partisan interests and logrolling come to the forefront, whereas full publicity encourages grandstanding and rhetorical overbidding. Conversely, secrecy allows for serious discussion, whereas publicity ensures that any deals struck are capable of withstanding the light of day." The force-of-the-better-argument norm, thus, rules out bargaining, but is vulnerable to "plebiscitory rhetoric". Nevertheless, Elster concludes, on the whole publicity "will lead to more equitable conflict resolutions". "If we take account of equity effects as well as efficiency effects, arguing in public is probably a superior form of collective decision-making than bargaining in private." There is nothing inherent in deliberative democratic theory which would contradict the proposition that transparency may lead to less flexible positions, even if this is a normatively highly problematic assumption for deliberative theorists. But while inefficient and time- consuming decision-making processes may be a disadvantage of transparency it is more of a side-line objection than a core objection for the deliberative idea of a civilizing effect. It is an unwanted by-product of publicity, but it does not break the causal chain of actors being forced first to talk the talk and then walk the walk. If actors are forced to shift rhetoric they will become entrapped by their own statements. Assuming that a decision will eventually be taken in one way or another the publicity of the process before that, this line of reasoning goes, will have set the agenda in such a way that policy options reflecting purely partisan interests will already have been forced off the agenda. If the parties eventually bargain an agreement from two public-spirited alternatives, then transparency has won the day. From a negotiation theory perspective, on the other hand, there would be no reason to oppose the proposition that if actors are forced to talk the talk in public, i.e. use arguments which adhere to widely accepted social norms, they will also subsequently have to walk the walk. To the contrary, this is precisely the meaning of positions assuming a higher degree of finality. Instead, much of the disagreement on whether transparency has a positive or a negative effect on the performance of political actors stems from the fact that while deliberative theorists emphasize effects on arguing and arguments, negotiation theorists concentrate on effects on positions. The issue of transparency reforms seems to put institutional reformers before a trade-off between efficiency, on the one hand, and decision-making processes characterised by arguing with public-interest justifications, on the other, while the two theoretical perspectives focus on one side each of the equation. However, this is not the end of the story. While this is hardly explicated in the literature the negotiation theory perspective also contains a more directly contrary objection to the deliberative story of publicity's civilizing effect. This objection states that transparency will increase, rather than decrease, the level of self-interested justifications. Thus, transparency will not make actors talk the talk in the deliberative sense, and therefore they will not walk the walk either. The background is that negotiation theorists make a completely different assumption, compared to deliberative theory, about who the public audience is. In deliberative democratic theory 'the public sphere' is a core concept. Public policy is legitimated in deliberations in the public sphere. This is where people meet as citizens, discussing the collective problems and public policy issues of the day. As citizens people are rational, critical and willing to deliberate over common interests and ideals. The proposed civilizing effect of transparency is derived from the assumption that the public audience, which decision-makers exposed to publicity will have to face, is an audience of such citizens. It is "a universal audience", in Manin's words, "characterised by the views and values commonly shared by all the citizens". In the negotiation theory models, on the other hand, the public audience consists of constituents. Negotiators are representatives: "Whether it is his employer, his client, his employees, his colleagues, his family, or his wife, every negotiator has a constituency to whose interests he is sensitive." A public audience of constituents, it is assumed, is neither public-regarding—there is no non-selfishness norm—nor interested in arguing over the best public policy. The constituents have already made up their minds about what they want. Preferences are fixed and solely self-interested and there is no or little willingness to compromise. It is such constituents that negotiators address when they act in public. And the negotiators' main concern at that point is to signal that they, as representatives, are on their constituents' side. Therefore, according to David Stasavage; "in any empirical investigation we should expect to see more uncompromising positions taken during open-door bargaining, greater polarization of debate, and more frequent breakdowns in bargaining then would otherwise be the case." Two contradictory hypotheses thus remain: 1) Increasing transparency will make elite political actors talk the talk, in the deliberative sense, which means using rational public- spirited arguments rather than bargain from self-interest. 2) Increasing transparency will put elite political actors—who are always representatives of some constituency—under pressure to emphasise the special interests of their constituents harder. In the empirical study which is presented below these two hypotheses are tested on representatives of companies and business associations acting to influence environmental policy decisions in Sweden and in the European Union. Comparing lobbyists backstage and frontstage In public, according to Thomas Risse, "even actors such as profit-seeking multinational corporations must justify their actions on the basis of some common goods or shared values". Organisations and groups with a positive "social construction", as Ann Schneider and Helen Ingram argue, have more political clout. "There are strong pressures for public officials to provide beneficial policy to powerful, positively constructed target populations and to devise punitive, punishment-oriented policy for negatively constructed groups." Hence, if the non-selfishness norm and the force of the better argument norm dominate public forums business lobbyists have to care. The methodological challenge for a study of effects of transparency and publicity is to get hold of data from the non-public backstage, which can be compared with similar data from the public frontstage. That problem was approached in two different ways in this study. First, rather than speaking to the main actors (and potential norm-breakers) themselves—company leaders and business association managers—I chose to talk to their guides to the political sphere. Interviews with the best, according to reputation, public affairs consultants in Brussels and in Stockholm was carried out. These consultants have as their job to help business leaders to success in the political sphere. Their unique skill, which makes them useful as interview objects, lies in being able to 'read' the incentive structure of the political arenas on which their clients must act. If there are strong social norms active in a particular context, they should know. The interviewed consultants were given a fictitious case study to comment on: A road transport organisation having a problem with a new draft proposal for eco-taxes on trucks, the story went, asks for their advice. The Secretary General of this hypothetical association goes to private meetings with officials and makes a public speech before a broad audience. The consultants are asked to explain how they would advise the Secretary General to act in the different settings. The degree of transparency and publicity was varied both in the switching of scenarios in the case study and in the comparison between the Brussels and the Stockholm consultants. The notoriously opaque European Commission, on the one hand, and the Swedish governmental ministries, with its centuries old institutionalised publicity principle, on the other, represent two extreme points on the transparency scale. The 'sunshine laws' in Sweden are exceptionally far reaching. Almost everything that is put on paper within government institutions is made publicly available. Even secret information may be legally leaked by civil servants to the media. The European Union, on the other hand, is regularly accused of lacking transparency. The European Commission has only recently acquired a freedom of information act, which was still in the process of being implemented during the fieldwork of this study (2001-2002). Officials who are caught leaking information may loose their job. Thus, if institutionalising transparency is an important determinant of political behaviour, there should be different signals coming from the Swedish ministries compared to the European Commission to those who wish to influence them. The public affairs consultants, if any, should be able to notice and interpret those signals. The interviews consisted of three parts. First, the consultants where asked to describe the content of the first meetings with the new (fictitious) client, the European/Swedish Road Haulage Association (ERA/SRA). Given the client's preference for getting rid of the eco-tax proposal, what would be the first things to analyse and sort out before the lobbying campaign could start? How would a typical client behave in these meetings and what would the role and input of the consultant be? Thereafter, the consultants were asked to advice the Secretary General of the ERA/SRA how to act in private meetings with the civil servants responsible for drafting the proposal, the Environment directorate of the European Commission (DG ENV) in the Brussels case and the Ministry of the Environment in the Stockholm case. How would the Secretary General act in order to make the civil servants abandon, or modify, the draft proposal? Which types of argument would he use? A list of nine arguments—three self- regarding and six public-regarding—was presented to the consultants, who were asked to make a selection for the private meeting. Finally, the same list of arguments was considered for a different context; a public speech at a big transport conference where 'everybody' (including policy-makers, media, representatives from competing modes of transportation, environmental NGO:s, as well as some of the ERA/SRA:s own members) was present. Would the ERA/SRA use the same types of argument in public as in private? Thus, while the two first parts of the interviews investigate the thinking behind a successful behaviour on the private backstage, the third part looks at how that behaviour would change at the public frontstage. The second way in which the methodological crux of being able to compare backstage to frontstage behaviour was approached was by comparing public and private documents. I was able to collect a sample of letters, which were sent off by industry lobbyists in Brussels to the European Commission under the presumption that they would remain confidential. These letters were compared with public position papers and press releases from the same Brussels organisations, on the one hand, and with letters addressed to the Swedish ministries, from similar types of senders, who know that all incoming mail will become publicly available, on the other hand. Thus, while the study of the consultants' guided tour to politics seeks to analyse the incentive structures for behaviour coming from the different institutional settings by means of a fictitious case study, the letters and press releases document real action. The letters study controls to what extent the consultants' story of what lobbyists should do in order to be successful corresponds to actual behaviour. While the comparison between Brussels and Stockholm analyses an institutional transparency effect the shifts between public and private action within the same institutional context will be referred to as a publicity effect. The presentation of the results is divided into two parts, following deliberative democratic theory's hypothesis of the transparency effect. First, the mode of communication is studied, i.e. the question of whether lobbyists are arguing or bargaining in private and in public. Secondly, the types of justification used in the different settings—self-regarding or public-regarding—are analysed. Arguing or bargaining? Before proceeding to the data it is necessary to clarify the distinction between arguing and bargaining. Deliberative theorists usually define bargaining as "a form of interest-aggregation that builds on the exchange of threats and promises", while arguing "is based on claims of validity". An actor who aims at convincing others of the merits of a certain policy option is arguing. If the communications are focused on reaching a common decision, but include no efforts at changing the minds of the other parties about what they want or what they perceive to be right, then bargaining is occurring. What was analysed in the empirical study, thus, was whether lobbyists, in public and in private, were trying to convince their counterpart of the merits, or lack of merits, of certain policy options, or whether they were trying to make them change positions in return for something—a positive compensation of some sort (promising/integrative bargaining) or the relief of a negative consequence over which the lobbyist has power (threatening/distributive bargaining). The interview study The public affairs consultants' story of the difficulties facing business leaders, who have their home base in the world of markets and consumers and who are forced on an involuntary visit to the world of politics and public opinion, was remarkably consistent and told in almost exactly the same way in Sweden as in Brussels. The first meetings between a public affairs consultant and a client typically revolves to a large extent around the problem (from the consultant's point of view) of getting the client to broaden his/her mind to the perspective of others. "You can't be so damned self absorbed if you are to be a good lobbyist". Business clients who are 'amateurs' in the political world, according to the consultants, tend to be egocentric, emotional, reactive and short-sighted. These characteristics risk leading to a hostile bargaining behaviour which is not acceptable by the political institutions—in Sweden and in the EU—and which will incur long-term political costs for clients. Egocentrism stems from the fact that the first reaction of any company or trade association to a political intervention within their field of business is 'how does this affect us?' Business people, according to the consultants, know their businesses, that is what they are used to focus on and care about. But they often have difficulties understanding that "what is important to them is not important to the rest of the world". "You do not manage a company like Volkswagen if you are not a very clever person. But you also do not get to the top of Volkswagen if not all your life has been focused on the car-market, if you're a sort of vague, broad-minded philosopher". Clients are also often emotionally involved in the issues. But industry representatives are not expected to be openly angry and emotional: "Policy makers switch off as soon as industry becomes emotional on jobs and starts hitting the table." An important part of the public affairs consultant's job, therefore, is to dampen frustrations. Amateurs also typically fail by being reactive, which implies being unable to come up with alternative solutions to the policy problems that the decision-makers are trying to solve. One consultant complained that the industry trade associations in Brussels "more often fight solutions than make solutions": They go like 'Oh, we welcome the White paper on transport, there was a real need for the Commission to talk to us about the future of the transport sector in the 21st century', which is the usual bullshit, and then 'but, we don't like this, we don't like that, we don't like blah blah'. With no alternative solutions! There is no 'This is a good idea, but it can be adapted like this and that'. Short-sightedness, finally, means making the fatal mistake of not considering the long-term relations with the political institutions. "You must remember that you never lobby on just one issue and that's it—'bye bye Brussels'. You'll be there for years and years and there will be more issues. You'll have to be on good terms with the institutions in order to make your points to them—that is absolutely key". "For the road hauliers it's best facing that they're in this for the long haul, not just the road haul". Egocentrism, emotionality, reactiveness and short-sightedness combined tend to lead to a "shotgun approach" of "let's kill this directive": 'BANG, this is going to ruin our industry, people's jobs, there are whole families out there doing these things'. And there's a hope, I wouldn't even say expectation, that this is going to put so much pressure on the decision-makers that they will say 'ok, let's forget about it'. But it's not going to work that way. The pressure approach antagonises and "breaks down the open dialog you want to be creating. You want to be seen as responsible and engaged. You want them to be aware of other solutions than eco-taxes". The appropriate role for industry lobbyists in their contacts with officials is that of the committed partner, who is using arguments that add value to the process and who seeks solutions which are acceptable for him/herself and for others. Moreover, those solutions should be argued for, not sold or hammered in. The lobby-corridor is not a market place for buying and selling policy proposals or pressuring and demanding. The dialogue "with which I think we would have the largest chance of success", as one consultant put it, is to "try to show that these proposals [the eco-tax] are not good for the environment". "When you walk into a meeting with commission officials—don't just try to get messages across. Don't talk too much. Listen to their concerns, ask them questions, try to have a two-ways discussion. When they have a question, surprise them by having all the data with you". The goal is to convince the civil servants that the draft proposal is a bad idea. "What are you trying to do? You are trying to change people's minds! The NGO's try to change the decision-makers' minds in the direction of an eco-tax, and I'm trying to change them away from that. That's the process". Therefore, the amateur characteristics must be subdued in clients before they enter the political arenas. The consultants' job in the first meetings is to help clients broaden their minds, cool off, be proactive and focus on long-term security rather than single battles. But while arguing is the regulative norm that does not mean that bargaining does not occur. The consultants are rational actors who calculate on the costs for their clients of norm- violation. If they believe it is in their client's interest to abandon the committed partner role and break the force of the better argument norm they will advice them to do so. When a client is loosing big, and the long term political costs related to norm violation are not judged to exceed the business costs involved with loosing, the committed partner may turn into a pressure group. "Somewhere you have to draw a line and say 'this is becoming too painful'". Pressure may then take the form of threatening with not being willing to cooperate in later implementation phases where industry assistance is important, as will be demonstrated in the letters study. But most often pressure implies involving lawyers. "Lawyers mean conflict". If that step is taken, then "you try to force them [the European Commission] down into negotiation by making the legal burden on them to justify their specific measure unbearable." Using threats and pressure is not just another lobbying tactic, however. When this happens it constitutes a breach of what is considered normal behaviour in the lobby corridor. The long term costs of such an action in terms of political credibility, according to the consultants, are high. The letters study The main conclusion of the letters study, with respect to the question of arguing vs. bargaining, is that the consultants' picture of a strong force-of-the-better-argument norm is confirmed also when it comes to real action. The content analysis was based on 58 previously confidential letters sent off to the European Commission, 55 publicly available letters addressed to the Swedish Ministry of the Environment and 41 press releases and public positions papers. The senders are all large companies and business associations whose businesses involve road transport, chemicals and other hazardous substances. All documents refer to environmental policy issues. Again, the question for the content analysis, with respect to the mode of communication, was weather these documents include attempts on behalf of the sender of convincing the receiver of the merits of different policy options by using rational arguments, or attempts to persuade the receiver to shift policy stance in return for a positive compensation or the relief of a negative consequence. If the closed lobby arena was a place for buying and selling public policy, confidential letters would include offers such as 'if you do X, we will give you Y,' perhaps followed up with 'and we can motivate it publicly in such- and-such a fashion.' If it was a naked power struggle we should see threats and demands. Almost nothing of those types of behaviour were found in the confidential letters, nor in the published and publicly available documents. Even with a very generous interpretation there are just ten paragraphs or sentences in the 154 documents (588 pages of text) which constitute potential instances of threats and promises. Most of these are either vague references to breaches of WTO agreements or offers of voluntary commitments, which is an established environmental policy instrument in the EU. Only three letters—all from the battery industry to European Commission officials—were as a whole dominated by a bargaining type of behaviour. The battery lobbyists were using both a positive offer and a negative threat in order to persuade the Commission to abandon any thoughts of a ban on Nickel-Cadmium batteries and to refrain from leaving the industry with the full costs of the collection schemes for all consumer batteries. The offer consists of a voluntary commitment, where industry takes responsibility for the collection and recycling of NiCad batteries. The threat is about non- cooperation in the collection and recycling schemes if a ban on NiCad batteries is introduced and if industry is burdened with the full costs for the collection of all consumer batteries. The most explicit threat was found in a letter from the European Portable Batteries Association, dated the 20th of June 2001, which was addressed to all commissioners. It included the following paragraph: In order to ensure that the environmental aims of the Directive can be satisfied, the measures therein must bear some relevance to industry realities. The success of implementation of the Directive is dependent on the participation of industry actors. You will understand that, despite our proven policy of support for environmental solutions, the Battery Industry will not be in a position to assume its part of the responsibility in the collection and recycling schemes, unless the establishment of financing tools is included in the proposal. [Emphasis as in the original]. This letter is qouted, however, because it illustrates the exception. Arguing is clearly the dominant mode of communication used by lobbyists, both in Brussels and in Stockholm, in public and in private correspondence. Not one letter, press release or public position paper contains a position on behalf of the sender without an argument backing it up. Just bringing 'wants,' without trying to justify them in some way, does not seem to happen. The phraseology, verbiage, and tone of the letters are polite and correct, well suited for a committed partner. Both the interviews and the letters show that lobbying, in private and in public, is about arguing rather than bargaining. Interestingly enough this result falls outside the predictions of both deliberative theory and negotiation theory, with respect to their hypotheses on the effects of transparency. The backstage-frontstage logic inherent in both hypotheses assumes that the private backstage is a place for bargaining. The implications are most striking for deliberative theorists, who put most emphasis on the arguing – bargaining distinction. The role of transparency in the deliberative version is to make actors shift from bargaining to arguing, but if they are already arguing behind closed doors this will obviously not be the effect! On the other hand, the fact that even such an opaque backyard of the political sphere as the lobby corridor is governed by the force of the better argument norm provides strong support for the general claim of deliberative democratic theory, that politics can be something more than a game of naked power. Although industry lobbyists in many respects are the complete opposite of ideal deliberative actors—they are willing to open their minds to the arguments and perspectives of others only because they need that information in order to promote their own interests—the fact that they have to adhere to the force of the better argument norm even at the backstage is a positive result from a deliberative view. Most regulated industry sectors could take a fight if they wanted to, for example by threatening with legal action, non- compliance or obstructing the implementation of regulations, but usually it is not in their interest to do so. These results suggest that deliberative theorists, when addressing the issue of transparency, has built a strawman of what goes on behind closed doors, which to some extent contradicts their own standard contribution to political theory. Types of justification The interview study From the list of nine arguments the consultants were asked to pick those three arguments they believed would be most useful for the European/Swedish Road Hauliers Association to use in private meetings with the civil servants responsible for the drafting of the eco-tax proposal. They were also asked which three arguments they would be least eager to use in that situation. Thereafter, the same list of arguments was presented to the consultants with the context changed. Now the Secretary General of the ERA/SRA was situated at a public conference, were he had been invited to hold a speech before a broad audience of policy-makers, people from the media, NGO:s, etc, as well as some of the ERA/SRA's own members. The point of letting the consultants pick and drop three arguments was not primarily to produce a quantitative assessment of the publicity effect, but rather to get the discussion on the quality of the different types of justifications going. Nevertheless, the selection illustrated a pattern which was subsequently confirmed and explained in the consultants' motivations of their choices. This pattern showed that self-regarding arguments, while not among the consultants' favourites in any setting, were more likely to be used in public than in private. On a scale from 45 (all consultants selecting the three self-regarding arguments) to -45 (all consultants dropping those three) the result for the self-regarding arguments was -24 in private but only -12 in public. While there was no significant difference between the Brussels and the Stockholm consultants' selections (i.e. no institutional transparency effect) the consultants were more willing to use self-regarding arguments for the public speech than in private meetings (publicity effect). This pattern contradicts the deliberative hypothesis that open doors will promote less self-interested rhetoric, but gives support, at least with respect to the publicity effect, to the negotiation theory hypothesis that publicity promotes putting emphasis on group interests. The self-regarding arguments are the least popular on the list for the backstage meetings, and fairly (although less) unpopular also on the public frontstage. The reason for that is not that there is a strong non-selfishness norm active. "Everyone's interest, within reason, is legitimate", as one consultant put it. In fact, the consultants argued, business lobbyists have to use these arguments in order to be seen as serious and trustworthy. Bringing only public- regarding justifications is simply not credible. As an industry representative you cannot pretend that you have engaged in the issue for other reasons than your own interest. Instead, the fact that the client believes that 'it will be a serious hit to the hauliers industry' should be clear from the beginning. That is the reason why the ERA/SRA is engaged and should be considered a legitimate partner in the dialogue. But it is not the reason why the decision-makers will change their minds about the eco-tax proposal. I'll be very clear about that [it'll be a hit to the industry] and have facts and figures, but I wouldn't dwell on it very much. It's the context, that's why I'm here, that's why we turned up today, that's the backdrop, but you don't need to go on about it for a long time. 'This is who we are, this is how many road hauliers there are,' that type of thing, costs and tax burdens, put it all on a couple of pages and point out that you have far more detailed information when and if they require it. On the backstage self-regarding arguments should be used for presentation—"if you never talked about your own interest, possibly you wouldn't be talking about your own qualifications for speaking to this audience in the first place. They would say 'well, why are you here?'" —but not for persuasion. Ideally, the message should be transparent with respect to the connection between the public interest and clients' private interests. "Link the two— show that it's good for me and for the rest of Europe—that's a winning argument". A public speech of the kind described in the fictitious case is a true public affairs challenge. The consultants perceived not a public audience, but several. The speech-writer of the Secretary General must take into account and find a balance between three different requirements, which all pull in different directions. First, there is a strong consistency constraint forcing actors not to deviate too much from the committed partner role played backstage. Those decision-makers who are present both at the backstage and the frontstage must recognize the message. They do not accept, according to the consultants, a sharp distinction between backstage and frontstage behaviour. I would use exactly the same arguments [in the public speech as in the closed meetings], because you will be singled out in two minutes if you are not living up to your messages. When you have defined the three things you want to say you do not move from that. You can adapt your points, slightly move them around in the debate, which shows intelligence, you can tailor your messages to different audiences—the way in which you will deliver them, the rhetoric around it—but not fundamentally the content of a message. You can stress one more than the others but you're not suddenly saying that it's a job issue, you're still saying it's a competition issue. The ground for your argument stays the same. Otherwise they'll see through you. Second, a speaker at a conference should for rhetorical reasons address the whole audience, however heterogeneous it may be. While the consistency constraint works against fundamental shifts in the selection of arguments public speeches require a different type of rhetoric than private meetings. There will be a certain degree of acceptance from those parts of the frontstage audience who have access also to the backstage, that in public speakers can not address only them. The Secretary General should try to define a broadest possible 'we.' "It is important to find something that everybody can be interested in". The effect of this perceived need to seek common ground is not, however, a change towards more public-regarding justifications. Rather, the attempt to address the whole audience, while at the same time holding on to the basic public-regarding message used in the private lobbying meetings, tends to result in a lower degree of specificity. The argumentation becomes vaguer and shallower, less focused on concrete decision-making. In the public speech the purpose is not primarily to change people's minds about the eco-tax; "you're not going to get the proposal revised there". Rather, it is an occasion to work on the industries' social image. The rhetoric and the style, the ability of the speaker to make the audience feel at ease—"to make sure that the speech is interesting and exciting, that the audience preferably applauds in at least two places" —is equally or even more important than the precise content of the arguments. "I am much more flexible when it comes to this type of occasion. A lot of this rhetoric [referring to the list of arguments] may be used". Third, the real challenge of the speech lies in balancing the consistency requirement and the requirement to be perceived as seeking common ground with the expectations of the ERA's/SRA's members. Depending on the internal situation within the association, the consultants argued, the leadership might have to include arguments which will 'make the members happy', in order to defend their own positions. The leaders of these types of organizations have a fairly difficult task. On the one hand, they have to keep their own comrades happy and 50 percent of their own group, and maybe more, demand blood: 'Tell that damned Environment Minister what a damned idiot he is'. That is a difficult balancing act, because he has to maintain a conciliatory tone toward the outside. If the Secretary General of the organization is up for re-election six weeks later, his tone toward the government may be a bit sharper. This is the main reason why self-regarding arguments are more likely to be used in the public speech than in private meetings with officials. In public, those arguments are for internal rather than external use. "In the speech he will say 'it will be a serious hit to the hauliers industry.' It is an argument that will not work on decision makers, but you put it in there to make the members happy. It is a way to unite your own people". The problem, from an external public affairs point of view, with putting weight on self- regarding justifications in the speech, is not that they violate a non-selfishness norm. The reason why the Secretary General is concerned about the eco-tax proposal—that it affects the hauliers' economic margins—is uncontroversial. The problem lies in holding on to a consistent performance compared to the backstage, where self-interested arguments should be used carefully and only for presentation. The letters study The letters study confirms the consultants' picture of how to prioritise different types of justification in public and in private. Self-regarding arguments are being used in the confidential letters, but not as frequently as environmental arguments. The number of documents containing self-regarding arguments is higher among the public documents than the confidential letters, contrary to deliberative theory's hypothesis and in line with the predictions of negotiation theory. The content analysis was based on simple counts of the number of letters including self- regarding and public-regarding justifications to back up the positions promoted by the senders. The basic unit of analysis was the letter. Three categories of public-regarding justifications were used—environmental (including public health and references to 'sustainable development') justifications, macroeconomic (including benefits for consumers) justifications, and references to undistorted competition (including free trade). The results are shown in tables two and three. Table 2. The institutional transparency effect: Confidential letters in Brussels compared to publicly available letters in Stockholm. Types of justification Stockholm (publicly available) Brussels (confidential) Difference (Transparency effect) Self-regarding 53% 33% +20 Environment 62% 76% -14 Macroeconomic 31% 33% -2 Competition 13% 22% -9 Note: Frequency of documents including at least one argument of respective category. 58 confidential (Brussels) and 55 publicly available letters (Stockholm) are used in the analysis. Table 2 shows the institutional transparency effect. The number of letters containing self- regarding justifications is higher in the publicly available letters in Stockholm than in the confidential letters in Brussels . The Stockholm lobbyist must calculate the risk that an especially interested public, including the media and whistleblowers among opponents, will ask for the letter. The difference in the degree of self-interest between the Stockholm and Brussels letters is probably—following the consultants' story—explained by the risk that the association members are among that especially interested public. The comparison between confidential letters and press releases and public position papers from the same Brussels organisations investigates the publicity effect. While table 3 shows that all types of justifications are used more in public than in private, the increase is largest for the self-regarding arguments. Table 3. The publicity effect: Confidential letters compared to press releases and public position papers from Brussels organisations. Types of justification Brussels (public) Brussels (confidential) Difference (Publicity effect) Self-regarding 71% 33% +38 Environment 80% 76% +4 Macroeconomic 66% 33% +33 Competition 34% 22% +12 Note: Frequency of documents including at least one argument of respective category. 58 confidential letters and 41 press releases and public position papers were used in the analysis. The fact that there is an increase in all types of justifications in public is also consistent with the consultants' story, which said that there is going to be a more careful selection of arguments in private communications than in public. Acting directly in public seems to allow more of an anything goes attitude. All the arguments available are used. Conclusions In order to predict the effect of transparency reforms and other efforts at increasing the publicity of political decision-making processes one needs to understand the behavioural logics of both the private backstage and the public frontstage. This article has demonstrated that with respect to business lobbying deliberative democratic theory makes the wrong assumptions at both arenas. The deliberative idea of a civilizing effect of transparency idealizes the frontstage, by overrating the public-spiritedness of the public and ignoring the dynamics of representation. But it also makes a strawman of what is going on at the backstage, by assuming that this is a place for self-interested bargaining. While business lobbyists could use both threats and promises if they wanted to, and sometimes do, it is normally in their best interest in private meetings with officials to abide by the force of the better argument norm. The mistake of assuming that bargaining is the rule behind closed doors is ironic, since the main argument of deliberative democratic theory in general is that politics is more about arguing and transforming preferences than rational choice theorists assume. Negotiation theorists also typically assume that the backstage is a place for bargaining rather than arguing. Nevertheless, the negotiation theory view of what transparency and publicity does to the behaviour of political elite actors was largely supported in the empirical study. Meeting the 'public' is not just a question of meeting the constituents—there is also cross- pressure from the requirement to address the whole audience and keeping a consistent message towards those parts of the frontstage audience who also have access to the backstage—but the constituents is an important factor. For business lobbyists it is important enough to push them towards more self-interested rhetoric in public. The fact that a broader audience requires a 'broader' argumentation implied more of plebiscitory rhetoric—a lower degree of specificity in negotiation theory terms—rather than more public-regardingness. How far beyond the European lobbying context will these results travel? A general conclusion is that models of two-level games are not enough to understand transparency effects on elite behaviour. Both deliberative theorists and negotiation theorists assume that the frontstage is one game, where the public consists of either public-spirited citizens or self-interested constituents, which is nested with the backstage. The public affairs consultants, however, instinctively felt that they were put before a trilemma, rather than a dilemma, at the frontstage. A public speech must strike a balance between consistency with the backstage, a broader public and constituents. To business representatives the pressure from constituents was felt harder, but other actors under different circumstances may prioritise differently. Where a given actor finds his/her optimal position in this trilemma will depend on the character of the particular backstage and frontstage, as well as on the actor itself and the issue at hand. Obviously, a representative of a decentralised trade union acting on an issue of high salience to its members will feel the constituency pressure harder than a leader of a more hierarchical organisation, such as a company, on an issue of low visibility. The degree to which the frontstage is characterised by strong public interest norms, on the other hand, will have to do with factors such as the degree of a common identity of the citizens and the integration and deliberative qualities of the public sphere. With respect to the backstage the strong presence of the force of the better argument norm ruled out bargaining as the dominant mode of interaction in the lobby corridors. The fact that even business lobbyists—the most market-oriented actors in politics—in private communications with officials who are working in a uniquely opaque political system (the EU) have to argue rather than bargain in order to promote their interests, and that the winning types of justification are public-regarding rather than self-regarding, is evidence of a strong deliberative democratic culture in Europe. Apparently, this deliberative culture does not require transparency to work. Whether we would find the same patterns in newly democratised countries or non-democracies is an open question. It is also not obvious that it holds for the American case. The famous European culture of compromise may be a factor here. Some consultants claimed that American companies "don't understand Europe", that they tend to be too confrontational for European standards, more prone for instance to threatening with legal action. "In Washington you could pursue a public affairs case where the best man wins, in Europe things don't work that way. [Here] you enter a dialogue rather than a fight." For Europe, at least, the conclusion must be that while transparency may have other benefits, such as strengthening accountability and 'input-oriented' legitimacy, with respect to the deliberative democratic virtues of arguing public-regardingly it does not seem to make political actors behave any better than they already do. Notes See, for example, Christopher Lord, A Democratic Audit of the European Union (Basingstoke: Palgrave, 2004), p. 220, Adrienne Héritier, "Composite democracy in Europe: the role of transparency and access to information," in Journal of European Public Policy (10:5, 2003), p. 824, Michael Zurn, ""Democratic Governance Beyond the Nation-State: The EU and Other International Institutions," European Journal of International Relations, (6:2 2000), p. 204. For discussions on the importance of transparency for international organisations in general, see Robert O. Keohane and Joseph Nye, "Redefining Accountability for Global Governance,", in Kahler, M. and Lake D., eds., Governance in a Global Economy: Political Authority in Transition, (Princeton: Princeton University Press, 2003), David Stasavage, "Open-Door or Closed-Door? Transparency in Domestic and International Bargaining", International Organization, (58:4, 2004). Fritz W. Scharpf, Governing in Europe: Effective and Democratic? (Oxford: Oxford University Press, 1999), p. 6. Fritz W. Scharpf, "Problem-Solving Effectiveness and Democratic Accountability in the EU." MPIfG Working Paper, (03/01, 2003), p. 4. Adrienne Héritier, "Elements of democratic legitimation in Europe: an alternative perspective," Journal of European Public Policy (6:2, 1999), p. 272. Roberto Gargarella, "Demanding public deliberation. The Council of Ministers: some lessons from the Anglo-American history", in Erik Oddvar Eriksen and John Erik Fossum, eds., Democracy in the European Union: Integration through deliberation? (London: Routledge, 2000), p. 202. Héritier 1999, p 271. Tony Bunyan, "Access to documents 'could fuel public discussion'", in Tony Bunyan., Deirdre Curtin and Aidan White, Essays for an Open Europe (European Federation of Journalists, www.ifj.org, 2000), p. 5. Cf. Stasavage 2004, p. 27f, Fiona Hayes-Renshaw and Helen Wallace, The Council of Ministers, (London: Macmillan, 1997), p. 7. Amy Gutmann and Dennis Thompson, Democracy and disagreement. Why moral conflict cannot be avoided in politics, and what should be done about it, (Cambridge, Mass.: The Belknap Press of Harvard University Press, 1996), p. 95. Roger Fisher, William Ury and Bruce Patton, Getting to Yes. Negotiating an agreement without giving in, (2:nd ed., London: Random House, 1999), p. 36f. Fisher, Ury & Patton 1999, p. 36. Jeremy Bentham, Political Tactics. The Collected works of Jeremy Bentham, in Michael James, Cyprian Blamires and Catherine Pease-Watkin, C. eds., (Oxford: Oxford University Press 1816 [1999]), p. 29. John Stuart Mill, Considerations on representative government, (London: Routledge, 1861 [1928]), p. 194. John Dryzek, Deliberative Democracy and Beyond: liberals, critics, contestations, (Oxford: Oxford University Press, 2000), p. 1. See, for example, Stephen Macedo, ed., Deliberative Politics. Essays on Democracy and Disagreement, (Oxford: Oxford University Press, 1999), Jon Elster, ed., Deliberative Democracy. (Cambridge: Cambridge University Press, 1998), James Bohman and William Rehg, eds., Deliberative Democracy: Essays on Reason and Politics, (Cambridge, Ma: MIT Press, 1997), Jurgen Habermas, Between Facts and Norms. Contributions to a Discourse Theory of Law and Democracy, (Cambridge: Polity Press, 1996). Joshua Cohen, "Deliberation and democratic legitimacy," in Alan Hamlin and Philip Pettit eds., The Good Polity, (Oxford: Blackwell, 1989), p. 19. Bohman & Regh 1997, p. xii. Elster 1998, p. 105. Cf. Thomas Risse, "'Let's Argue!' Communicative action in world politics," International Organization, (54:1, 2000), p. 32. Macedo 1999, p. 10. I am referring here primarily to Thomas C. Schelling, The Strategy of Conflict (Cambridge; Mass.: Cambridge University Press, 1960), Richard E. Walton and Robert B. McKersie, A behavioural theory of Labor Negotiations. An analysis of a social interaction system. (Ithaka, NY: ILR Press, 1965), Robert D. Putnam, "Diplomacy and domestic politics: the logic of two-level games," International Organization, (42:3, 1988), Fisher, Ury & Patton 1999. Also more recent game theoretical works support the concerns of classical negotiation theory with respect to transparency and publicity. Tim Groseclose and Nolan McCarty, who have modelled the impact of an audience in bargaining games, conclude that publicity increases the risk for negotiation breakdown and Pareto inefficient outcomes. "Although there may be benefits to 'sunshine laws' and other measures to make negotiations open, our results show that they may actually harm efficiency" (Tim Groseclose and Nolan McCarty, "The Politics of Blame: bargaining before an audience," American Journal of Political Science, (45:1, 2001), p. 114. Similarly, Stasavage argues that public posturing "can provoke a break-down in bargaining that has a negative impact for all concerned" (Stasavage 2004, p. 673). Walton & McKersie 1965, chap. 4. Walton & McKersie 1965, p. 159. Walton & McKersie 1965:93. Cf. Schelling 1960, p. 28. Walton & McKersie 1965, p. 94. Elster 1998, p. 117. Simone Chambers, "Behind Closed Doors: Publicity, Secrecy, and the Quality of Deliberation", Journal of Political Philosophy, (12:4, 2004), p. 393. Jon Elster, "Strategic uses of arguments", in Kenneth Arrow, et.al., eds., Barriers to conflict resolution, (New York: Norton, 1995), p. 258. Jon Elster, Alchemies of the mind. Rationality and the emotions, (Cambridge: Cambridge University Press, 1999), p. 402. If deliberation implies reaching agreements by transforming preferences, which is the usual definition, and if publicity at the same time reduces actors' willingness to change their positions, then 'public deliberation' seems to be something of a contradiction in terms. Bernard Manin, "On legitimacy and political deliberation," Political Theory, (14, 1987), p. 358. Fisher, Ury & Patton 1999, p. 49. Apparently negotiators are also assumed to be men rather than women. Stasavage, 2004, p. 679. Risse 2000, p. 22. Ann Schneider and Helen Ingram, "Social construction of target populations: Implications for politics and policy," American Political Science Review, (87:2, 1993), p. 334. Reputation was captured simply by talking to a lot of people within the lobbying communities in Brussels and in Stockholm (consultants, trade association representatives, companies, civil servants, researchers). Not a foolproof method of finding 'the best' of course, but pragmatically reasonable. The sample consists of the directors of public affairs or environmental affairs of eight consultancy firms in Brussels and seven in Stockholm. Torbjörn Larsson, "How open can a government be? The Swedish experience", in Veerle Deckmyn and Ian Thomson, Openness and transparency in the European Union, (Maastricht: European Institute of Public Administration, 1998), pp. 34-36. See, for example, Christer Karlsson, Democracy, Legitimacy and the European Union. (Uppsala: Acta Universitatis Upsaliensis, 2001), p. 64, Richard Bellamy and Dario Castiglione, "The uses of democracy. Reflections on the European democratic deficit", in Eriksen & Fossum 2000, p. 69. The three self-regarding arguments read: "The new proposal for an environmental classification system on trucks should be withdrawn because 1) … it will be a serious hit to the European [Swedish] hauliers industry 2) … the system encourages cheating at vehicle inspections, which will be difficult to control. 3) ... it is unreasonable to sacrifice the hauliers like this, when it is obvious that the sources of air pollution are complex and thus are a responsibility for the whole society. The public-regarding arguments read: 1a) … it will distort competition between the different modes of transportation, which will impede on the implementation of an effective common market for freight transportation. [Brussels], 1b) … measures against air pollution requires co-operation on a European or global level, not unilateral Swedish rules [Sweden]. 2) … a vehicle tax is an ineffective means to reduce emissions from the transport sector. 3) … the industry employs a lot of people directly and indirectly – jobs will be lost. 4) … the European [Swedish] hauliers industry needs less trouble and cost burdens, not more. 5) … the European [Swedish] economy is dependent upon letting companies have access to road transports at reasonable prices. 6) ... the heavy focus on road transports prevents the search for a comprehensive approach to air pollution, which has many sources, ultimately hindering effective environmental protection. The new regulation on access to documents from the EU institutions (EC NO 1049/2001) came into force in December 2001. As I inquired into getting access to third party letters being sent to the Commission it turned out that the new regulation was interpreted, by the Commission itself, to apply retroactively to letters sent off before the regulation was even being considered. Zurn 2000, p. 192. Stockholm Interview Person No 5 (IP 5). Stockholm IP 6. Brussels IP 5. Brussels IP 8. Brussels IP 2. Brussels IP 4. Brussels IP 8. Brussels IP 8. Brussels IP 8. Stockholm IP 5. Brussels IP 8. Brussels IP 5. Stockholm IP 2. Stockholm IP 2. Brussels IP 2. Brussels IP 6. Brussels IP 8. Brussels IP 6. Brussels IP 5. Brussels IP 2. Stockholm IP 6. Brussels IP 4. Stockholm IP 6. Stockholm IP 7. Stockholm IP 1. Brussels IP 1. See Robert D. Putnam 1988, George Tsebelis, Nested Games. Rational Choice in Comparative Politics, (Berkely, University of California Press, 1990). Stockholm IP 6. 1