Resumé

The original concept of the negotiated economy from the late 1970s provided some heuristic insights into significant developments of political, economic, and administrative processes. The later development of the concept has not brought the theory of the negotiated economy any further. This is a reply to an article published in SPS (Andersen et al. 1996) which responded to criticisms brought by me in an earlier article (1994:4). The defence of the theory of the negotiated economy is not consistent. The theory is both non-positive and positive, and normally accepted criteria for empirical confrontation are not adhered to.

Introduction

In Scandinavian Political Studies (1996:2) Niels A. Andersen, Peter Kjær & Ove K. Pedersen responded to an article in an earlier issue of the journal (Christiansen 1994:4) in which I criticized some of the work based on the concept of the negotiated economy. In this rebuttal I briefly comment on their main points.

The concept of the negotiated economy originates from the Norwegian power study where it was heuristically used to denote the phenomenon that market transactions were increasingly "circumscribed" by negotiated solutions (Berrefjord 1978, 155). In the original version, the concept was not further developed theoretically. The concept in its original somewhat loose meaning was later used fruitfully in a few empirical studies (e.g. Midttun 1988; Berrefjord & Hernes 1989). In the later - mainly Danish - versions, the concept of the negotiated economy was developed into a theory. Briefly, my criticisms were that (1) the "second generation" of analyses based on the negotiated economy are theoretically unfruitful, (2) that the theory is not useful for empirical analysis, and (3) that it is normatively biased. In this rebuttal I concentrate on the first two points. As regards the third point of my criticism, I sustain my previous points to which I have nothing to add.

The defence of the theory of Andersen et al. is not consistent and even
adds to the confusion about what the negotiated economy is about.

Is There a Theory of the Negotiated Economy?

With regard to my theoretical critique, Andersen et al. claim that I have not only misrepresented but have not fully understood the core of their theory. The latter point is true. In their reply to me and elsewhere it is simply not possible to arrive at a clear understanding of the theoretical status of their theory.

According to Andersen et al., ". . . the negotiated economy is not to be viewed as a theory in the conventional sense. We do not wish to claim that the term negotiated economy is a theoretical representation of a positive phenomenon (i.e. a stable, objective, and observable phenomenon) ..." (p. 169) (emphasis in original). This interesting position is substantiated by two arguments that contradict what they were supposed to support. The first argument is that "... the concept of negotiated economy is an attempt to capture a particular logic of organization which we claim has emerged in a particular historical context in Denmark" (p. 170). The second argument is that "we have never claimed that this institutionalized logic is the only logic of organization in Danish society . . ." (p. 170). In both cases, the defence of a non-positive theory is substantiated by referring to positive phenomena! It is not possible to have a non-positive theory and at the same time claim that the theory somehow mirrors a positive phenomenon.

Andersen et al. defend their position by referring to some kind of "constructivist" methodology (p. 167) which is apparently in opposition to what they claim is my "non-constructivist" position (p. 167) or "liberal social ontology" (sic!) (p. 174). They probably mean some kind of crude empirism. According to my knowledge of the social sciences, nobody today defends a pure positivist ontology or epistemology. It never becomes clear by what means their "constructivistic" approach allows them to make a nonpositive theory refer to a positive phenomenon.

Even as a purely theoretical construction, the theory of the negotiated economy has a number of significant problems. The theory does not define or apply precise concepts. It is very difficult to come to grips with the concepts applied as well as the basic theoretical claims. For instance, in a couple of publications it is argued that the theory encompasses phenomena based on communicative, strategic, as well as instrumental rationality (Nielsen & Pedersen 1988, 211-218; Berrefjord et al. 1989, 33-39; see also Nielsen & Pedersen 1988a, 95-100). Besides the theoretically unclear difference between strategic and instrumental rationality and a basic uncertainty about the theory's conception of actors, it is a major problem

to develop a theory in which actors in some relations pursue an individually maximizing strategy and in others respond to communicative rationality. The actors of the theory are motivated by directly opposite motives! Such a theory inevitably runs into problems concerning falsifiability. In the end, the theory is able to explain everything and, thus, nothing (King et al. 1994, lOOff.). A number of other minor and major problems relating to the theory of the negotiated economy could be mentioned. Core concepts are not defined, relations between variables are unclear and formulated in a way that precludes precise assessment. It is often unclear how conclusions are reached. In sum, assessed on its purely theoretical qualities, the theory disqualifies itself on some of the basic criteria for good theories. Thus, if the proponents of the negotiated economy see fit to stick to the claim that the theory does not somehow represent a positive phenomenon, their contribution is insignificant and uninteresting. It is like a piece of soap in the bath tub. It is hard to grasp, and when you do - it slips away again.

As has already been shown - and as any interested person will learn from reading the different contributions of the theory - the proponents of the negotiated economy are actually interested in saying something about the political, social, and economic structures of present society. Therefore, assessing the theory by its fruitfulness for understanding or analysing empirical objects is inevitable.

Is the Theory of the Negotiated Economy Testable?

Of course, not all social science theories are or can be subject to empirical
analysis. However, if a theory is to be used as some kind of a positive theory
it is worthless, if it cannot be exploited for empirical analysis.

Concerning the status of empirical analysis in the negotiated economy, the contribution by Andersen et al. is very confusing. In the first half of their reply it is claimed that "... the later macro-theoretical concept of the negotiated economy cannot be tested empirically" (p. 168), ". . . although the concept as such may not be tested empirically there are still good reasons to maintain it as an analytical lead" (p. 168), "We do not have an empirically verifiable theory of the negotiated economy" (p. 171), and ". . . most of the concepts we use to organize social experience have never been and can never be tested" (p. 172). A few lines further on, this cannonade is followed by the statement that "... we still think that empirical analysis is the key to determining the value of a particular theoretical lead or strategy of analysis" (p. 172)! Even constructivists cannot eat their cake and have it too!

Andersen et al. seem confused about the meaning of empirical testing. In
the social sciences, theories or concepts are never tested as such. Concepts
and theories are not constructed as pure and full representations of "reality"

(King et al. 1994, 49). Their very meaning is to condense or formalize aspects that from one point of view or another are found analytically interesting. It is meaningless to test empirically a concept or a theory in that sense. When we test theories, we mean that we test implications of the theory. From theory A we deduce that X implies Y and not Z. A strong test of the theory implies that if we find Y, theory A will be strengthened, and we will be able to falsify the competing theory B. Through careful operationalization we find a valid expression of X, Y, and Z and set out for empirical analysis. The logic is as simple as that. The problem - and empirical testing is a world full of problems - is of course to make valid deductions from the basic theoretical claims and to make valid tests of the carefully deducted hypotheses.

It is unclear how the proponents of the negotiated economy pursue an empirical strategy. Among the many "second generation" publications on the negotiated economy it is difficult, if not impossible, to find examples of a clear and well-conducted empirical study respecting normally accepted criteria for empirical research. Their own references to"a quite substantial, and expanding, body of empirical studies based on the second generation concept of negotiated economy" (p. 172) include Pedersen et al. (1992) and Andersen (1994) among others. I am not going to comment in detail on these studies. As empirical studies they are all more or less flawed. Pedersen et al. (1992) is a strange example of an empirical study. Its method seems to be to collect examples in support of their theory instead of systematic empirical analysis. The problem of falsifiability is never considered. They jump to conclusions and forget to account for the reasoning concerning the relations between theoretical claims and empirical results. The results often contradict simple observations. Of course it is always gratifying to see social scientists arriving at results that contradict common sense or established viewpoints. However, we must demand that it is made clear how these results are reached. This implies accounting for data and analytical method. It is one of the fundamentals of scientific research that results can be communicated and shared among scholars (King et al. 1994, 7ff.), if not for any other reason then at least to find out precisely what disagreements are about. The proponents of the negotiated economy seldom present accounts that make it possible to follow their way of reasoning.

Another study (Andersen 1994) is only loosely coupled to the basic theoretical claims and largely unrelated to the claims of the negotiated economy. Still, it suffers from the same fundamental lack of precision as to its database and its method of analysis.

Again, these criticisms cannot be rejected by referring to some principle of constructivism or by claiming that their critics are non-constructivist. There are basic demands which we cannot and should not give up (King et al. 1994, passim.). Among these are accepted rules of reasoning, the demand that

scientific methods and results must be communicated and thus made public,
and the demand that empirical references reveal a basic understanding of the
phenomenon studied.

My own study of public regulation of the manufacturing sector (Christiansen 1993, 1994) and the criticisms of the theory of the negotiated economy were based on a survey of the relations between firms and public authorities in a number of regulatory areas. The study may not be very good or representative, but at least it is clear how conclusions are reached. Briefly put, I deduced two hypotheses from the basic claims of the negotiated economy. Neither of the two hypotheses were supported in confrontation with data based on a representative survey of Danish manufacturing firms. I never claimed that my test has general validity. However, I do claim that the "second generation" theory of the negotiated economy is in serious trouble. My study does not cover all relevant areas. Andersen et al. find that I ". . . reduce() firm-environment relations to a question of public regulation" (p. 172). Of course, this is not so. Their allegation underlines the lack of understanding of the basic logic of empirical tests. The point is that public regulation of firms is a significant area of interrelations between the market system and the political system. If it is impossible to find general traits of negotiated economy in this area, the theory is significantly weakened. It would be gratifying if the proponents of the negotiated economy could show one piece of empirical analysis in support of their theory. So far, they cannot.

The Role of Non-Hierarchical and Non-Market Coordination

The observations of increasing inseparability of the political and the market system were what made some of the scholars in the Norwegian power study invent the label "negotiated economy". Hernes (1978, 2lff.) focused on how the ideal types of democracy, bureaucracy, and market increasingly have been infused with logics from the other two types. As mentioned already, Berrefjord (1978) used the negotiated economy to identify this phenomenon. The same phenomenon has been noted by other observers. Scharpf puts it this way. "There is more coordination in the modern world than is plausibly explained by the classic mechanisms of community, market, hierarchy and their commonly discussed variants" (Scharpf 1994, 27).

There exists a whole body of positive phenomena in relation to which it might be very useful to exploit concepts to characterize non-market and nonhierarchical coordination. To mention just a few examples: in most Western countries a number of previously traditional public services - typically in areas where natural monopolies motivated the running of a state business company - have been transformed into governing systems resembling those

of private business. These companies are seldom left over to pure market governance. For different reasons some kind of public governance is upheld (Horn 1995, 134-169; Grøndahl & Grønlie 1995). Outsourcing is another example of the use of market mechanisms in order to exploit the gains of market exposition while at the same time maintaining political control of the service produced. Some branches of industry, such as shipbuilding, which are exposed to the competitive pressures of the market and at the same time heavily dependent on public regulation and/or subsidies are other cases in point. Studies of such branches might reveal how the logics of different systems of governance influence the strategies of actors in the private as well as public sector.

These phenomena - and many more - are all characterized by institutional arrangements that are neither pure market, nor pure hierarchy. It is a highly worthwhile job to come to grips with some of these institutions. Actually, a couple of studies based on the concept of the negotiated economy are available. However, they do not represent the "second generation" of the negotiated economy. Berrefjord & Hernes (1989) have produced a nice piece of analysis on the role of large firms in present society. Midttun's (1988) paper on Norwegian hydropower plants is an interesting study of a sector characterized by non-hierarchical and non-market means of coordination.

Much more analysis is needed if we are to come to a better understanding of some of these phenomena. If we were to do that we would have to go through three logical steps. The first step is to delimitate the object. This might be a difficult and lengthy business, but the task is unavoidable. The second step is to develop and apply clear theoretical concepts in order to understand some of the essentials of the object studied. Parenthetically, it should be noted that the large and growing body of new institutionalist literature could provide a strong theoretical frame for such analysis (cf. Williamson 1985; Ostrom 1986; Knight 1992). The third logical step is to apply these theoretical concepts to empirical analysis. So far, the "second generation" of studies of the negotiated economy has not fulfilled even one of these steps.

The recent analyses based on the negotiated economy seem to be a science studied by only a small group of individuals. It is a science that is difficult to relate to other theoretical developments in a precise way, and the international literature has shown but scant interest in these theories. In a review article, Jordan & Schubert (1992, 24) find that the negotiated economy is "... inadequately grounded in the prior literature". There may be good reason to exploit the concept of the negotiated economy. If that is to be done, we are still in the in the initial stages.

REFERENCES

Andersen, N. ń Kjær, P. & Pedersen, O. K. 1996. "On the Critique of Negotiated Economy",
Scandinavian Political Studies 19, 167-177.

Andersen, N. Å. 1994. "Danmark: Forvaltningspolitikkens udvikling", pp. 29-63, in Lægreid,
P. & Pedersen, O. K., eds., Förvaltningspolitik i Norden. Copenhagen: Jurist- og
Økonomforbundets Forlag.

Berrefjord, O. 1978. "Fra embetsmannsstat til embetsmannsstat?", pp. 134-159, in Hernes, G.,
ed., Forhandlingsøkonomi og blandingsadministration. Bergen, etc: Universitetsforlaget.

Berrefjord, O. & Hernes, G. 1989. "Storforetak, marked og byråkrati", pp. 171-190, in
Nielsen, K. and Pedersen, O. K., eds., Forhandlingsøkonomi i Norden. Copenhagen:
Jurist- og Økonomforbundets Forlag.

Berrefjord, 0., Nielsen, K. and Pedersen, O. K. 1989. "Forhandlingsøkonomi i Norden - en
indledning", pp. 9-40, in Nielsen, K. & Pedersen, O. K., eds., Forhandlingsøkonomi i
Norden. Copenhagen: Jurist- og Økonomforbundets Forlag.

Christiansen, P. M. 1993. Det frie marked- den forhandlede økonomi. Copenhagen: Jurist- og
Økonomforbundets Forlag.

Christiansen, P. M. 1994. "A Negotiated Economy? Public Regulation of the Manufacturing
Sector in Denmark", Scandinavian Political Studies 17, 305-319.

Christiansen, P. M. 1995. "Forhandlingsøkonomi. En kritisk vurdering", Tendens. Tidsskrift
for kultursociologi 7, 22-34.

Grøndahl, 0. N. & Grønlie, T. 1995. "Frihet og styring - et uløseligt problem?", pp. 201-231,
in Grøndahl, 0. N. & Grønlie, T., eds., Fristillingens grenser. Bergen: Fagbokforlaget.

Hernes, G. 1978. "Makt, blandingsøkonomi og blandingsadministrasjon", pp. 11-60, in
Hernes, G., ed., Forhandlingsøkonomi og blandingsadministration. Bergen, etc.: Universitetsforlaget.

Horn, M. 1995. The Political Economy of Public Administration. New York: Cambridge
University Press.

Jordan, G. & Schubert, K. 1992. "A Preliminary Ordering of Policy Network Labels",
European Journal of Political Research 21, 7-27.

King, G., Keohane R. O. & Verba, S. 1994. Designing Social Inquiry. Scientific Inference in
Qualitative Research. Princeton: Princeton University Press.

Knight, J. 1992. Institutions and Social Conflict. Cambridge: Cambridge University Press.

Midttun, A. 1988. "The Negotiated Political Economy of a Heavy Industrial Sector: The
Norwegian Hydropower Complex in the 1970s and 1980s", Scandinavian Political Studies
11, 115-143.

Nielsen, K. & Pedersen, O. K. 1988. "Fra blandingsøkonomi til forhandlingsøkonomi. Mod et nyt paradigme", pp. 171-226, in Klausen, K. & Nielsen, T. H., eds., Stat og marked. Fra Leviathan og usynlig hånd til forhandlingsøkonomi. Copenhagen: Jurist- og konomforbundets Forlag.

Nielsen, K. & Pedersen, O. K. 1988a. "The Negotiated Economy: Ideal and History",
Scandinavian Political Studies 11, 79-101.

Ostrom, E. 1986. "An Agenda for the Study of Institutions", Public Choice 48, 3-25.

Petersen, J. 1989. "Pensionsfonde som aktieejere - Nye institutionelle muligheder for samfundsøkonomisk forhandlingsstyring på det danske kapitalmarked", pp. 243-269, in Nielsen, K. & Pedersen, O. K., eds., Forhandlingsøkonomi i Norden, Copenhagen: Jurist- og Økonomforbundets Forlag.

Scharpf, F. W. 1994. "Games Real Actors Could Play. Positive and Negative Coordination in
Embedded Negotiations", Journal of Theoretical Politics 6, 27-53.

Williamson, O. 1985. The Economic Institutions of Capitalism. New York: The Free Press.