Harald Saetren: Iverksetting av offentlig politikk: En studie av utflytting av statsinstitusjoner fra Oslo 1960-1981. Universitetsforlaget: Bergen, 1983. 208 p.

Last autumn Harald Saetren defended his doctoral thesis on "Implementation of public policy: A study of the relocation of state agencies from Oslo 1960-1981". Together with Pär-Erik Back (Umeå), I had the stimulating task of being an official opponent on the thesis from Bergen. In this review I intend to summarize some of the points made during the dissertation.

Perhaps a more adequate subtitle for the study would have referred to the lack of real relocations of state agencies from Oslo. The last year of the study ends the official attempts to relocate agencies, attempts which had been on the agenda for some twenty years. And the results are meagre. Only a few hundred jobs were moved outside Oslo during the period and altogether only about 2000 new state administration jobs were created outside the capital region. At the same time within Oslo about 15,000 new jobs were created in the state sector. According to some statistics, the number of state institutions within Oslo increased from about 180 in the early 1960s to 290 in the mid-19705.

During this period the government had twice put the matter up for general discussion in the Storting and had twice received approval for its policy. Twice the government had itself tried to carry through this policy. Both times the policy itself had been formulated by governmental commissions. Against this background one may understand the focus of Saetren's study. He asks what went wrong. How is it possible that so little came out of so much decisionmaking activity?

True to the Bergen tradition within public administration research, Saetren starts with some polemical comments on the traditional ways of trying to grip the problem. He wishes to advance three perspectives based on organizational theory that run counter to assumptions within what he labels traditional theory. It should be noted, in passing, that it is not altogether clear what this traditional perspective is and who the exponents of that view are.

The three assumptions that he challenges are 1) that decision-makers always and foremost are interested in substantial results, 2) that the implementation process is crucially different from the decision-making process and 3) that the implementation is always a product of the will of goal-conscious actors.

These assumptions are challenged by introducing three different pespectives on implementation. Within the first perspective, implementation is seen as symbol-politics. Participants in decision-making can be interested in the participation itself or they may be trying to legitimize their position through participation. It is in no way necessary for this participation to be directly

related to substantial results regarding relocation of agencies. Symbolic
decision-making processes are known by their preoccupation with general
principles.

The second perspective sees implementation as interest group politics. Implementation is viewed much the same as any decision-making. Interested parties are negotiating and trying to have their way regardless of decisions that have already been made. Interested parties form coalitions. They have different resources and they are in different strategic positions. All in all, this perspective views implementation very much in terms of traditional pressure-group analysis. Implementation failure can be interpreted in terms of new actors entering the arena or in terms of changes in the resources of the actors. Perhaps we could also talk about changes in interests, at least if we are looking at decision-making over decades.

The time perspective seems to be very important for the third perspective. Within this social by-product perspective a number of things are hinted at, and I cannot confidently say that I have grasped the full meaning of the perspective. In any case, implementation can be influenced by many unexpected changes in the decision-making environment. The actors themselves have only limited capacity to concentrate on different questions. One day one thing commands their attention, another day it is something else. Unforeseen changes may take place calling for a new strategy. Decisions made elsewhere may influence and solve at least part of the problem at hand.

In the Allison tradition, Saetren holds that probably no single perspective
alone can help us understand what went wrong. The perspectives complement
each other and together they allow for deeper understanding of the process.

The main bulk of the book is of course a case study reporting how the decision-making process advanced chronologically. We are given a very good account of the general socio-economic situation giving rise to the interest in relocation. After the main chronological part we also find a good analytical chapter in which the entire process is seen through the eyes of different categories of actors. And all this leads us up to the final conclusions of the study. For those conversant with perspectives similar to those advanced by Saetren it comes as no surprise that the symbol-perspective aids us in understanding the interaction between the government and the Storting and between both of these and the electorate. In short, the symbol-perspective accounts for the cooptive rhetorics of the politicians. The ways in which the concerned agencies tried to overcome governmental intentions is intepreted within the interest group perspective. The last perspective seems to catch what the two other perspectives have missed. Important developments took place during the decision-making and implementation process. What originally was a major problem did not appear as such twenty years later.

The study provokes a number of interesting questions. The interpretation

given in the study seems to lean heavily on the limitations made. These beg some questions. Is it clear that we are really faced with an instance of implementation? Could it perhaps be contended that only a few final decisions were made and that in many of these cases agencies were in faet relocated? What seems to be crucial here is that only general guidelines were asked for from parliament. Neither could the government agree on concrete decisions producing clear guidelines for the agencies. One could ask therefore to what extent there exists a clear authoritative decision-making for the cutting point between decision-making and implementation.

Another question deals with the substantive limitations made. How are we to know that the actors were not in fact concerned firstly with regional policy and only secondarily, as one among possible means, with relocation? In many places in the study it is mentioned that the government in fact tried to keep many options open and that it proceeded with many other measures to alleviate regional problems. In fact the argument to abandon relocation altogether was announced in just this way. The relocation policy was to be resumed only if other alternative measures failed. Some evidence suggests that such other measures were perhaps effective, as for instance the measure to put a ceiling on the growth of public employees in Oslo.

The reason that I bring up these questions is of course that the answers may point toward a much more traditional interpretation of the results. If few binding decisions were made and if relocation was only one among many alternative means, perhaps a traditional rationalistic perspective on decisionmaking would have sufficed. These points could therefore have been much more thoroughly dealt with.

Another interesting point concerns the complementarity of the perspectives. Could it be that they are mutually exclusive, at least to some extent? Let us assume that the symbol model is correct. According to this model, the implementation failure should be understood as a consequence of the decisionmakers' lack of interest in the matter, as a consequence of the actors being interested in other matters than relocation. Consequently the implementation lacks a motor, it lacks clout. If this is so, what is left for the other perspectives to explain? The victory of the interest groups would be a victory over nothing. And it seems to me that similar arguments could be made regarding the other perspectives. For this reason I would have liked to see an attempt made to integrate the perspectives into one framework, suggesting crucial variables for future use — we need to predict as well as to understand.

The problem of the thesis suggests that Saetren is surprised by the implementation failure. But do we in fact have reason for surprise? Could it not be argued that success would have been more surprising, taking into account the type of policy we are dealing with? We have a redistributive policy. The administrative establishment is challenged. The goods to be redistributed

can only be given to a few, but there are any number of potential receivers. Those from whom something will be taken are strong and homogeneous, whereas their counterparts are divided and weak. Political logic alone would seem to dictate that a government would not be all that committed to relocation.

Against this background we may be surprised indeed that a fairly sizable relocation programme was carried through in Sweden at the same time as the programme failed in Norway. Harald Saetren discussed this in one of his dissertation lectures. He used the comparison with Sweden as a means of strengthening some of the observations made in his case study on Norway alone. The difference in strength between a case study and a two-country comparison may not be all that large. The comparative design is still plagued by the problem of overdetermination.

During the dissertation I tried to make use of an additional case, Finland. Here we find a decision-making process which fits the Norwegian one almost perfectly. This pertains not only to the type of political and other actors involved but also to the level of ambition. The results were, however, even more futile than in Norway. At the same time, Finland has many similarities with Sweden. The very preliminary comparison that I tried to make seems to suggest that Saetren stresses a number of explanatory factors too much. Among these we find the resources given to the commissions preparing relocation proposals, the character of the policy proposal itself (many relocation proposals that are explicitly made with reference to where and when), the extent to which the government is taken hostage by the administrative interests, the shortage of space for agencies in the capital, etc.

On the other hand, there is one area of inquiry that may be worth a more detailed scrutiny than the one given to it by the author. The Finnish case and indeed the Swedish one would point to the importance of party-political factors. These are for some reason not made very explicit in the study. It may indeed be correct that they are inconsequential in Norway, but still a number of passages in the thesis point to their potential importance. I have a hunch that in this respect much more could have been done with the inteviews made by the author with a large number of ministers in all those cabinets of different political colour that sat during the reform period.

As the reader hopefully has seen from my discussion of the thesis, it is a work of great interest. The main case study offers rich insight into a long and very complex decision-making process, and the perspectives used to organize the bewildering variation give rise to many intriguing questions. This to me is one of the hallmarks of a good study.