Resumé

This article analyzes the implementation of two regional development programs. The purpose
is to show how the programs approved by the Riksdag were translated into practice and to
explain why the funds were distributed as they were. We seem to have found substantation
for the hypothesis that the borderlines between politics and markets are blurred in this
implementation process. Bargaining was an important element in the process. Decisions were
shaped in interaction between authorities and companies and, at the central level, in the
actual process of bargaining. The most important finding is that the main conflict in this
bargaining economy was not between the decision-makers and the company. Rather, it was
between the central and regional levels.1

Mixed Administration and Bargaining Economy

On 8 April 1981 'Regional Day' had come round again in the Swedish Riksdag. One after the other, members seized the opportunity to draw attention to the growing problem of unemployment in their own regions — Norrbotten, Värmland, Bergslagen, Landskrona, Göteborg ... This show of local sentiment was inspired by a debate on the regional policy bill (prop 1980/81:100: 17). A very large sum of money — over a billion Swedish kronor — was to be made available, in an attempt to halt migration from the already sparsely populated regions and to bring the employment situation in the forestry districts into line with the metropolitan areas. Funds were allocated to several programs, all of them geared to the achievement of regional balance. The implementation of two of these will be analyzed below. The process which was initiated by the Riksdag's confirmation of two of the programs can serve to illustrate the integration between politics and markets that appears to characterize the Western democracies of the 1980s.

It has often been claimed that in the present mixed economies of the Scandinavian countries there are strong connections and interdependencies between the market on the one hand and politics and administration on the other. Boundaries become blurred. Companies and their organizations appear as political actors, seeking to influence the formulation of public policies.

At the same time the market economy is undermined. Corporate decisions about production, employment and investment are actually made outside the market — by the government, the parliament or the administration, or by management in interaction with tråde unions and various political bodies. The state has assumed a large measure of social responsibility; it is not the market that alone decides what production should be maintained, what should be produced, or what investments should be made.

The division of roles between parliament and government, the administration, various organizations and companies is often a key factor in theorybuilding in the social sciences. The Norwegian sociologist and power researcher, Gudmund Hernes, has coined the phrase the mixed administration to describe a system in which companies and organizations take part in formulating public policy — an integration with respect to democracy and administration — and the bargaining economy to describe a situation in which the state and various organizations both affect the market. The state and the companies have become interwoven (Hernes 1978; NOU 1982:3. See also Berrefjord 1982; Eriksen et al 1980). Charles E. Lindblom, American professor of economics and political science, paints a similar picture in his survey of different political-economic systems. Lindblom emphasizes in particular the influence of the large corporations on politics, and his book "Politics and Markets" closes with the frequently quoted words: "The large private corporation fits oddly into democratic theory and vision. Indeed, it does not fit" (Lindblom 1977, 356).²

The Swedish situation, too, seems to be characterized by a measure of integration between politics and markets. We know that companies and their organizations take an active part in the formulation of public policy. The government and the Riksdag frequently make decisions intended to correct the workings of the market. The government intervenes in a variety of ways. Production and investment decisions_—possibly in an industry, a company or a production unit — are affected by decisions arrived at outside the companies. Every day we see the government using public funds to restructure industries, and we hear of companies, factories and regions being helped to survive as a result of government intervention. Public subsidy programs create new companies and new production units.

We can thus recognize the signs of blurring of the borderlines also in the Scandinavian countries.' So far social science research has produced little in the way of results, or at least of exhaustive results, in this field. In the present paper I hope to add something to current knowledge of the integration of public and private decision processes. In particular my aim is to illustrate integration as it appears in a specific policy area. The implementation of two regional policy programs can serve to illustrate the active role which companies play in the implementation of selective assistance programs.

The Swedish policy-making process is widely regarded as evincing a high degree of planning, thoroughness, rationalism, openness and consensus. The Swedish decision process has been described by Tom Anton as "extraordinarily deliberative, involving long periods of time during which more or less constant attention is given to some problem by well trained specialists. It is rationalistic, in that great efforts are made to develop the fullest possible information about any given issue, including a thorough review of historical experiences as well as the range of alternatives suggested by scholars in and out of Sweden. It is open, in the sense that all interested parties are consulted before a decision is finally made. And it is consensual, in that all decisions are seldom made without the agreement of virtually all parties to them". The result, claims Anton, is "decision making which never seems rash, abrupt, irrational, or indeed exciting" (Anton 1969, 94).

Some traits that are particularly characteristic of the Swedish policy-making process have probably given rise to these descriptions.⁴ There is for instance the public commission system, and the system of circulating material for comment (remiss-procedure). As a result of these two procedures the government and the Riksdag can generally assume that an issue has been thoroughly discussed by the time a decision is finally made. The commissions gather and analyze information, and they make policy recommendations. Reactions to questions before the government are collected from the authorities concerned. Various organizations with an interest in the matter often play an important part here, at both the investigatory and the consultative stage. Another element in the situation is the principle of free access by the public; the documents on which any decision will be based are available to public scrutiny, and the pros and cons of different alternatives are often debated. The policymakers will normally have collected material from various mutually independent sources, and will generally have access to investigations of alternative preconditions and alternative consequences of any measures under consideration, and the material on which a decision will ultimately be based may well include a number of opposing views on any or all of these aspects. When, after this often lengthy decision process, a decision is finally made, it is often founded on a broad consensus between the parties concerned and has generally the support of the relevant major organizatipns.

Pluralism is regarded as a positive and desirable element in the policymaking process that precedes a decision in the Riksdag. However, once the Riksdag's decision has been taken, pluralism is no longer considered appropriate. According to the accepted theory, decisions should then be implemented without outside interference. All political decisions are made by the government and Riksdag, and decisions reached at the political level are transferred to the administration for implementation. The administration is the servant of the body politic. It is operated by non-political civil servants

who obey the political bodies and translate their intentions into practical action. At this stage other actors exert no real influence. Influence stems from the political top and passes down through the administrative hierarchy; its direction is top-down, not bottom-up.

However, we all know that this is not how political decisions are implemented; studies of the policy implementation process have revealed quite a different picture. It can certainly not be taken for granted that policy programs will be implemented in accordance with the intentions of the politicians. Control is not a one-way affair, and the implementation process is a complex one. The authorities whose task it is to implement the programs are exposed to demands and pressures from a variety of interested parties. The formal decision of the authorities is very often preceded by (lengthy) bargaining between various actors, all of whom have their own interests to watch.⁵

I am going to describe the proceJs leading up to the receipt by the individual company of regional development assistance in the shape of location aid (depreciation loans and location loans). I will show how the programs approved by the Riksdag were translated into practice. The subject of my investigation is the process between the Riksdag's appropriation decision and the authorities' decision to grant aid. The programs to be studied here included depreciation loans of 225 million kronor and location loans of 600 million kronor, (prop 1980/81:100:17 81 ff). In fact, however, only 589 million kronor were paid out to companies in depreciation and location loans.⁶

The network of contacts of the decision-making body will be described below, and some idea given of the substance of the different contacts. For what purpose did the officials handling the case get in touch with the companies concerned? And what did the companies have to say when they made contact with the decision-makers? When we are investigating the presence — or otherwise — of a mixed administration and a bargaining economy, such questions must be asked in this context.

The present paper also has an explanatory purpose, and the question of why the funds were distributed as they were will also be discussed. Why were the decisions of the authorities formulated as they were? I shall try to analyze the deliberations of those concerned and to identify the factors that were taken into account in the decision-making.

The Riksdag's 1979 decision on regional policy implied some major changes in regional assistance policy. There were to be fewer types of assistance and there were to be changes in their technical composition. But regional aid was still to be concentrated to central and northern Sweden. In recent years most regional development aid has been assigned to the northern part of the country (Dsl 1981:15).

The present decision procedure has been in operation since 1 July 1979,

when the system of decentralized decision-making which had been in use on a trial basis since 1976 was confirmed. Thus, for the budget year 1981/82 decisions on regional development aid were made at three levels, by the County Administration, the National Labour Market Board and the government. In the latter case, applications had normally already been reviewed by the County Administration and the National Labour Market Board. The level at which a decision was made depended in part on the size of the aid applied for, and the kind of activity in which the applicant was engaged (SFS 1979:632). Although decentralization was fairly far-reaching in terms of the number of decisions, applications involving the largest sums of money were still handled at the central level (Dsl 1981:28).

Documents such as commission reports and other material germane to the decisions were used in the present study, but an important part of the investigation consisted of standardised interviews with the people handling regional development aid. The officials concerned in the Ministry of Industry, the National Labour Market Board, the County Administrations, the Regional Development Funds, and the County Employment Boards were interviewed. Any picture of the decision-making processes and the deliberations of the relevant authorities based only on the available public records would have been superficial and incomplete, as the reader will understand from the following account.

As we have seen, it is the northern regions which have received most
assistance in recent years. The County Administrations elsewhere in the country
have had extremely few cases to handle.

The interviews with officers of the Regional Development Funds, the County Employment Boards and the County Administrations were confined to the four northernmost administrative districts. The numbers interviewed were thus as follows:

The 22 interviews were held in the last quarter of 1982. During a conversation lasting 1-3 hours the officials answered a questionnaire containing 19 questions. Notes were taken during the interviews and they were then summarized together with the interviewer's general recollections of the meeting in a written "interview report" made immediately afterwards.

One further point should be mentioned. It is not only companies producing

goods and services which can be granted location aid; even local authorities may be entitled to assistance. However, I have limited myself to decisions connected with applications from companies, as my purpose has been to investigate the integration of private and public decision-making processes.

Government and Riksdag Establish the Rules of the Game, Leaving Liberal Scope for Interpretation

Once the Riksdag had reached its regional policy decision on 8 April 1981, what happened? The government received an official notice (riksdagsskrivelse), and then in turn issued an official document (regleringsbrev) to the authorities concerned on the conditions and the purposes of the appropriation (rskr 1980/81:233). Sometimes these documents may represent a fairly detailed form of control on the part of the government vis å vis the authorities, but this was by no means so in the case of location aid during the budget year 1981/82. On the contrary, the regleringsbrev cannot be said to have restricted the authorities' freedom of action in any way; all they contained was a statement of the resources which were being made available.⁷

There are thus no grounds for describing the implementation of the two regional development programs in hierarchical terms. In implementation literature the inadequacy of the allocated resources is generally cited as an obstacle to smooth implementation in accordance with the intentions of the politicians. Or there may be a shortage of officials to handle the matter or of competent personnel. And there may be considerable financial restrictions (Sabatier & Mazmanian 1979). However, in the present case there does not appear to have been any shortage of money. The permitted financial limits were wide enough to exclude any such possibility. The allocations were constructed in such a way that the stipulated amounts could be exceeded. The authorities were not compelled to establish priorities. Thus there should not have been any deviation from the intentions of the politicians because of lack of resources.

Vague stipulations as to how the authorities are to act can also be an obstacle to smooth implementation (Bardach 1977; Pressman & Wildavsky 1973). It is not unusual for laws and enactments providing rules and guidelines for the use of economic means of control to be formulated in vague terms. Guidelines for the allocation of the Riksdag's appropriation were already laid down before the Riksdag decision of April 1981. They were established in the ordinance on regional development aid passed in June 1979.⁸

The profitability and employment situation of the applicant company are
emphasized in the general conditions to be considered: 'Aid is granted only
to such operations as are judged to be achieving satisfactory profitability and

to be providing those employed with permanent employment, and which are now being conducted, or shall in future be conducted, where the conditions are appropriate to such operations' (SFS 1979:632 § 15, my italics). Risks connected with profitability were to be weighed against employment goals. However, the actors concerned might judge the facts differently, and they might also disagree on the weight they attached to these judgements. And even if the actors did agree in their understanding of the facts, they could still attach different weight to profitability and/or employment aspects. The ordinance with its vaguely formulated rules left plenty of latitude for different deliberations.

Further, consideration of an application should include an evaluation of the amount of assistance that might be granted. It should not exceed the sum required for an investment to be effected. This meant, first, that the decisionmakers should assess the government's contribution in light of the company's economic situation and alternative sources of financing, and secondly that they should assess the amount of subsidy required for the investment to come about (§§ 21, 29). Thus here, too, the decision-makers had to make evaluations and decisions about suitability. At the same time there were certain very specific rules in the regional development ordinance which limited the scope of the decision-makers, but these were connected exclusively with the amount of the aid, which was given in terms of maximum limits which only the government could alter (§§ 22, 29).

Thus, although the Riksdag specified the rules of the game, there was still plenty of latitude. Apart from the detailed regulations about the size of the allocations, the directives were vague and they left the authorities considerable room for manoeuvres. Assessment of the importance of the employment factor was left entirely to the administrative body concerned. Attitudes to risk-taking would naturally vary, and this could of course result in deviations from the intentions of the politicians.

A third obstacle in the implementation process may have to do with the officials concerned. Presumably those who handle the aid itself want to apply the rules in accordance with the intentions of the politicians (Sabatier & Mazmanian 1979). But here too there is plenty of latitude, since the stipulations in the ordinance give little guidance, for instance, as to how the profitability requirement and the desire for lasting employment opportunities should be weighed against one another. The impression that the implementation process is weakly controlled by the politicians is reinforced.

When the Riksdag makes an appropriation, it would also be possible for it to stipulate in detail hoy/ the resources which are being made available should be allocated. In our example such a procedure would have implied that the government and the authorities distributed the allocations between regions, industries and companies.

Another strategy, in complete contrast with this last, is to let actors outside the political system steer the implementation of the regional development aid programs, as was done in the present instance. The Riksdag laid down the rules of the game and indicated the conditions which should be fulfilled before an allocation could be made. From then on it was the companies' demands for assistance that decided the allocation between different regions, industries and companies.

If implementation was really to follow the intentions of the politicians, it was necessary that there should be companies in the process of making investments plans, and that these companies took the initiative and applied for aid. The decision-makers — the County Administrations, the National Labour Market Board and the government — were to remain more or less passive in the initial stages of the implementation process, waiting for applications to come in.₉ Thus we have here a fourth obstacle to implementation in accordance with the intentions of the politicians: the collaboration of actors outside the administrative bodies was required before anything happened.

The Actors and the Information: A Long and Complex Decision Process

Once the government had formulated its regleringsbrev and the National Labour Market Board had effected any changes in its instructions, reactions from companies were awaited. The decision process began when they responded by submitting an application, either to the County Administration in the district where they intended to locate their operations or to the National Labour Market Board.¹⁰

Applications for location aid were handled by, among others, regional bodies, the National Labour Market Board and the government. At the regional level responsibility was divided between the County Administration, the County Employment Board and the Regional Development Fund. The County Administration was the coordinating unit, and was responsible for the decision process and the decision at the regional level (SFS 1979:632). The expertise of a great many other public bodies was also called upon at this stage, depending on the type of application.

The Regional Development Funds examined the projects submitted, mainly
on a basis of economic estimates. The Fund assessed each project and then
decided whether a government subsidy was likely to make it profitable.

The County Employment Board examined the companies' applications
in light of the labour market situation in the district in question. The Board
also scrutinized the recruitment plan that was drawn up by the company

together with the County Employment Board itself, the relevant employer
organizations and the trade unions.

Starting from the pronouncements of the Development Fund and the County Employment Board, the County Administration then either reached a decision or referred the matter together with the comments to the National Labour Market Board for further processing. During the budget year 1981/82 the right of decision of the County Administration was limited, and many issues were to be decided at the central level.

The National Labour Market Board therefore made the supplementary investigations that were thought to be necessary. The Board made use of the available material on questions of industrial policy, which described among other things the demand and capacity situation in different industries, by getting the views of the National Industrial Board and other relevant bodies. The formal decision was made by the Director General of the Board on the recommendation of the officer dealing with the matter in the Location Section. By this time the recommendation would have passed through the working group within the Eioard's steering committee which consisted, apart from the Director General, of one representative from LO (the Swedish Confederation of Trade Unions), one from TCO (the Swedish Central Organization of Salaried Employees), one from SAF (the Swedish Employers' Confederation) and one member of the Riksdag.

If the matter was one which, according to the stipulations of the ordinance, had to be reviewed by the government, the National Labour Market Board submitted the application to the government, with comments, where it was investigated by the Ministry of Industry (SFS 1979:632 § 44). Additional information and personal contact with the applicant companies were not normally required at this stage on the scale necessary for the investigations of the regional bodies and the National Labour Market Board. Nevertheless there were informal discussions in almost all cases between the officers of the Ministry of Industry, the National Labour Market Board and the Regional Development Funds. In this way the Ministry was also able to form some idea of various aspects not normally mentioned in written pronouncements. By this time, however, the decision base was already both extensive and comprehensive, having been gathered from contacts with a great many different people.¹¹ Among these, of course, were many people outside the public sector. The decisions which were actually made by the Minister of Industry, Nils G Åsling himself, either concerned matters which fell outside the scope of the government ordinance, or they referred to cases involving some doubt. But all decisions were rooted in a long process of bargaining.

The companies took part in the decision process in direct contacts with
various private and public bodies to whom the matter was submitted for
comment. Thus all those interviewed from the National Labour Market

Board, the County Administrations, the County Employment Boards and the Regional Development Funds stated that in almost all cases they had been in direct contact with the company concerned. Often it was the company which got in touch with the authority. Even if the company did not seek such contact, the officials generally felt that they would like to hear its views.

Visits were often made to the company, or representatives of the company called on the authority concerned, in order to present their case. Sometimes it was a question of a formal attendance on the Minister of Industry or the Director General of the National Labour Market Board. In such cases, representatives of the regional authorities were often included, and sometimes also local members of the Riksdag.

Not infrequently the contacts at the central level led to a more penetrating exchange of views and to bargaining. The parties to the negotiations would be the officials of the authority concerned and the company's management. Sometimes a third party would be involved, namely the banks. If a representative of the company's bank was not actually present, the discussions nevertheless included the banker as an invisible guest at the table.

In this long and complex decision process the trade unions also played a part. The unions' views were generally expressed in a variety of ways, at both the central and the regional levels. The trade unions should have been informed already before a company submitted an application to the County Administration or the National Labour Market Board. Among other things they helped to draw up the recruitment plan for confirmation by the County Employment Board. When an application was being considered at the regional level, the parties on the labour market were consulted together. The National Labour Market Board referred the matter to the central organizations. The central trade unions were represented on the steering committee at the National Labour Market Board which assessed applications before the formal decision was taken by the Director General of the Board. The local trade unions also often took part in any formal attendance at the central level.

Decision-Makers' Deliberations: Decisions Shaped in the Course of Bargaining

Why were the funds distributed as they were, and why were the decisions
formulated as they were? In this section we shall look at the way that just
these decisions emerged from the process of discussion and bargaining.

Part of the explanation of the way funds were divided as they were among assistance areas, regions, industries and companies lies, as we have seen, in demand on the part of the companies themselves. No attempt at all was made by the decision-makers to steer the use of the funds, other than by applying

the very general guidelines indicated in the bill and the ordinance. The long
and complex decision process which preceded a formal decision was actually
set off by applications submitted by companies.

Formally speaking, questions of location aid were decided during the budget year 1981/82 by the County Administration/Governor, the Director General of the National Labour Market Board, and the government. In fact the County Administration's decisions appear to have been hammered out in a process of joint consultation between representatives of the County Administration itself, of the relevant County Employment Board and the Regional Development Fund. During this discussion process the stance was shaped which subsequently became the official view of the County Administration.

At the central level the picture looks quite different. At this level bargaining took place between officials of the National Labour Market Board/ Ministry of Industry and the applicant companies, resulting in agreements which later became formal decisions. The reason for this difference between the regional and central levels was probably that the decision-makers at the two levels were moved by different considerations.

I have already mentioned the evaluations which the decision-makers were to make according to the ordinance. But what considerations did the authorities actually take into account, when they assessed the applications received? The decision-makers were guided mainly by the following considerations:

1. Are these operations entitled to aid according to the conditions given in
the ordinance?

2. How does this project affect the structure of the industry and the company's
competitors?

3. Has management the capacity and qualifications necessary to be able
to implement the project?

4. Will the project be economically profitable?
5. What employment effects will the project have?

6. How large a government subsidy should the company receive?

Regardless of who was going to make the decision — the County Administration, the National Labour Market Board or the government — the question of entitlement to aid was always investigated. If the decision-maker decided that the company was so entitled, the way in which the investment would affect the structure of the industry was then examined. It was a question of deciding whether aid might distort the competitive situation. At the same time the possible benefit of launching a new technique could be taken into consideration, or the advantage of generally diversifying industry in the region. Considerations of this kind seem, however, to have weighed very

much more heavily at the central level than in the regions: they were rarely
mentioned as being important in our interviews at the regional level.

Management's capacity and qualifications for implementing the project seemed to be regarded as a very important consideration. In almost all the interviews this was mentioned as the needle's eye through which every application must pass. If the manager or managers concerned were not already known to the decision-makers, they would try to find out something about them preferably through personal contact.

We have already noted that the Regional Development Funds had to investigate whether projects were likely to prove satisfactorily profitable. The Funds based their investigation on economic estimates. The question was then investigated as to whether government loans and contributions could mean that the project in fact became economically profitable. This investigation provided the chief basis for the decision-makers' economic deliberations.

We have also noted that the main task of the County Employment Boards was to assess the possible creation of lasting employment opportunities. The deliberations of the decision-makers were based on these findings, and there were some calculations in which the pronouncements on employment weighed heavily. But in this respect the regional and central levels differed from one another, particularly as between the County Administrations and the National Labour Market Board.

More material naturally accumulated as a matter was referred upwards in the hierarchy; new and more detailed investigations were added to the store. In some cases the National Labour Market Board may not have agreed with the regional bodies on matters of fact. But above all the central bodies seem to have evaluated the various facts differently, choosing to give less weight to the employment goal. They were less prepared to take risks. Officials handling location aid in the Ministry of Industry have declared that the County Administrations find it difficult to fix the amount of aid appropriate to the individual case: "In most cases the County Administration seems to settle on the largest possible grant within the limits laid down in the ordinance" (Dsl 1981:28, 92). The regional bodies always seem to give priority to the employment goal, even if it involves very high costs.

The decision-makers also had to decide how large a subsidy — loans and/or other contributions — was needed to make the project profitable. At the regional level the amount of regional development aid was rarely questioned. On the contrary: companies were often encouraged to apply for the maximum subsidy allowed for each additional job opportunity as given by the National Labour Market Board in its guidelines¹² — and sometimes even far more. On this point the companies could normally count on the full support of the regional body when an application was being handled at the central level.

The regional body followed up the applications of "its" companies at the central level. Since the regional body — and the County Administration in its formal decision or pronouncement — accepted maximum grants, there was seldom any latitude for bargaining with the company. Thus the decisions could be reached unanimously.

At the central level, however, bargaining seems to have been started (the parties had conflicting interests) when the National Labour Market Board and the government in line with the intentions of the politicians tried to minimize government loans and contributions (SFS 1979:632 §§ 21, 29). The National Labour Market Board and the government paid attention to other considerations compared with the County Administrations.¹³

Thus at the central level bargaining took place between the authority and the company, and it could happen that a single matter became the object of bargaining in both the National Labour Market Board and Ministry. In almost all the issues handled at the central level bargaining-type situations arose, mainly to do with the amount of the possible aid, as the companies had applied for the maximum amount allowed for by the ordinance. These discussions could assume all kinds of forms, from telephone calls to direct talks at the Board or Ministry or at the company.

It seems as though the decision-makers — the National Labour Market Board or the government — wanted to "purchase" the job opportunities as cheaply as possible. Thus when it appeared to be a case of a serious company applying for aid, the direct contacts would lead to bargaining. It was then clear that the companies' goals had been to obtain the largest possible amount from the government, while putting in as little as possible financially themselves. Usually the two parties were able to reach an agreement.

This agreement was then normally approved by the Director General of the National Labour Market Board and the government respectively; the amounts settled on could not then be altered without renewing negotiations with the company.

The "number employed" was a constant variable in the bargaining process. The planned project was expected to generate a certain number of employment opportunities and this amount did not change. The bargaining was almost exclusively concerned with the size of possible government loans and/or other contributions.

According to the politicians, the purpose of regional aid is to help companies to grow. As to the considerations which weighed most heavily with the decision-makers, however, we can perhaps conclude that short-term employment goals were ascribed much more importance than any considerations connected with ideas of industrial expansion, i.e. new technology and new ideas, future growth industries etc. In all essentials the deliberations of the decision-makers were built up round their evaluation of the employment

situation. It is employment opportunities that the state is "purchasing". And
this emphasis on short-run employment goals appears to be strongest at the
regional level.

Conclusion

At the beginning of this paper I referred to our very limited knowledge of the way public and private decision processes are interwoven. We barely know what the integration consists of. After examining the implementation of regional development aid, we seem to have found substantiation for the statements of mixed administration and bargaining economy. The results can be summarized briefly as follows.

Bargaining was an important element in the implementation process. The politicians laid down the rules of the game, but left considerable latitude for interpretation; preliminary decisions were shaped in interaction between authorities and companies and, at the central level, in bargaining. Thus the regional and central levels differ considerably.

The decisions of the County Administration gradually emerged from the interplay between the County Administration itself, the Regional Development Fund, the County Employment Board and the company; the parties do not appear to have had conflicting interests. The companies normally applied for the maximum aid allowed according to the stipulations of the ordinance. Since the regional authorities — and the County Administration in its formal decision or pronouncement — accepted maximum grants, there was not usually any scope for bargaining with the company.

At the central level the actors were moved by quite different considerations, generally leading to bargaining between the National Labour Market Board/ the government and the company; the decision-makers at the central level did not accept maximum aid, and here the parties appear to have had conflicting interests. Bargaining started when the National Labour Market Board and the government, in line with the intentions of the politicians, tried to minimize the state contribution. The companies' goal was to obtain as much as possible from the state and to put in as little as possible themselves. The Board or the Ministry's negotiators were aiming for the maximum contribution from the company and its bankers, and the minimum amount in government subsidies. The bargaining was concerned with the amount of the possible aid; the number of employment opportunities was given.

Our most important finding therefore seems to be that the main conflict was not between the decision-makers and the company. Rather, it was between the central and regional levels. Against the central actors — the government, the National Labour Market Board, the National Industrial

Board, the central trade unions — were ranged the regional actors. These included as well as the company also the County Administration, the County Governor, the Regional Development Fund, the County Employment Board, the local authorities and the local trade unions. The regional and local authorities and the local trade unions supported the company and formed a regional coalition in the game. The regional bodies normally followed up the applications of "their" companies at the central level, seeking to bring pressure to bear on the central decision-makers. It was a game between the central and the regional levels, a game in which the companies had different roles to play.

---

NOTES

1. This article is a slightly revised version of a paper presented at the ECPR workshop on "Politico-Economic Interdependence in the Highly Industrialized Democracies", Freiburg, March 20-25, 1983.

2. The integration of politics and markets is often stressed in the literature. See for example Grant 1982, Okun 1975, Olson 1982. The integration in the implementation process is analyzed in Hjern & Hull 1982, Hjern & Porter 1981.

3. See, for example, Lundgren & Ståhl 1981, Riksdagens revisorer 1982, SOU 1981:72, Henning 1983. The integration of public and private decision processes in an individual case is described and explained by Henning 1980. See also Berrefjord 1982.

4. There is an extensive literature describing Swedish public policy-making in similar terms. The picture is fairly uniform. See, for example, Castles 1976, Hancock 1972, Meijer 1968, Richardson 1979, Ruin 1982, Tomasson 1978.

5. The importance of bargaining is often stressed, particularly in American literature. See for example Brigham et al 1980, Elmore 1980, Ingram 1977.

6. National Labour Market Board, Anslagsframställning för budgetåret (Appropriation report) 1983/84.

7. See Statsliggaren 1980/81 (Register of appropriations), Ministry of Industry.

8. From the budget year 1982/83 a new ordinance applies which should give the County Administrations wider powers of decision. But nothing seems to have altered in the way cases are handled, except that the County Administration is now able to make more decisions. See SFS 1982:677.

9. For views on the importance of external actors see, for example, Elmore 1978. The Regional Development Funds, however, played an active role.

10. Unless otherwise stated, the contents of this and the following section are based either on the individual answers of the officials interviewed or on summaries of their answers.

11. This picture of the handling of the location issues also appears in Dsl 1981:28 ch 5.

12. National Labour Market Board, Råd och anvisninger för länsstyrelsernas handläggning av ärenden om lokaliseringsstöd (December 1979).

13. This bargaining between the authority, the company and the bank are analyzed in Dsl 1981:28.

REFERENCES

Anton, J. 1969. "Policy-Making and Political Culture in Sweden", Scandinavian Political
Studies 4, 88-102.

Bardach, E. 1977. The Implementation Game: What Happens After a Bill Becomes a Law
Cambridge and London: The MIT Press.

Berrefjord, O. 1982. Rafnes-saken. Forhandlingsøkonomien i aksjon. Bergen: Universitetsforlaget.

Brigham, J. and Brown, D.W., eds. 1980. Policy Implementation. Penalties or Incentives?
Beverly Hills and London: Sage.

Castles, F.G. 1976. "Policy Innovation and Institutional Stability in Sweden", British Journal
of Political Science 6, 203-216.

Dsl 1981:15 Regionalpolitiskt stöd. Effekter? Effektivitet? Stockholm: Liber.

Elmore, R.F. 1980. "Backward Mapping: Implementation Research and Policy Decisions",
Political Science Quarterly 94, 601-616.

Elmore, R.F. 1978. "Organizational Models of Social Program Implementation", Public
Policy 26, 185-228.

Eriksen, 0., ed. 1980. Industri og politikk. Bergen: Universitetsforlaget.

Hancock, M.D. 1972. Sweden. The Politics of Postindustrial Change. Hinsdale: Dryden
Press.

Henning, R. 1983. "Industrial Policy or Employment Policy?" in Richardson, J. and Henning,
R., eds. Managing Unemployment. Policy Responses of Western Democracies. London: Sage
(forthcoming).

Henning, R. 1980. Partierna och Stålverk 80. En studie av industripolitiskt beslutsfattande.
Stockholm: Liber.

Hernes, G. ed. 1978. Forhandlingsøkonomi och blandingsadministrasjon. Bergen: Universitetsforlaget.

Hjern, B. and Hull, C. 1982. "Implementation Research as Empirical Constitutionalism",
European Journal of Political Research 10, 102-115.

Hjern, B. and Porter, D. 1981. "Implementation Structures: A New Unit of Administrative
Analysis", Organization Studies 2, 211-224.

Ingram, H. 1977. "Policy Implementation through Bargaining: The Case of Federal Grantsin-Aid",
Public Policy 25, 499-526.

Lindblom, C.E. 1977. Politics and Markets. The World's Political-Economic Systems.
New York: Basic Books.

Lundgren, N. and Ståhl, I. 1981. Industripolitikens spelregler. Stockholm: Industriförbundets
förlag.

Okun, A.M. 1975. Equality and Efficiency. The Big Tradeoff Washinton DC: The Brookings
Institution.

Olson, M. 1982. The Rise and Decline of Nations. Economic Growth, Stagflation and Social
Rigidities. New Haven and London: Yale University Press.

Meijer, H. 1968. "Bureaucracy and Policy Formulation in Sweden", Scandinavian Political
Studies 4, 103-116.

NOU 1982:3 Makutredningen. Slutrapport. Oslo: Universitetsforlaget.

Pressman, J.L. and Wildavsky, A. 1973. Implementation. Berkeley: University of California
Press.

Richardson, J.J. 1979. "Policy-Making and Rationality in Sweden: The Case of Transport",
British Journal of Political Science 9, 341-353.

Riksdagens revisorer. 1982. Det statliga stödet till industrin, Rapport 1981/82:3. Stockholm:
Riksdagens revisorer.

Ruin, O. 1982. "Sweden in the 19705: Policy-Making Becomes More Difficult" in Richardson,
J. Policy Styles in Western Europe. London: George Allen & Unwin.

Sabatier, P. and Mazmanian, D. 1979. "The Conditions of Effective Implementation: A Guide
to Accomplishing Policy Objectives", Policy Analysis 5, 481-504.

SOU 1981:72 Att avveckla en kortsiktig stödpolitik. Stockholm: Liber.

Tomasson, R.F. 1978. Sweden: Prototyp of Modern Society. New York: Random House.