Resumé

The three Scandinavian countries have been able to radically decentralize their public sectors without losing the ability to control macroeconomic performance. The explanation of this puzzle is that, in all three countries, institutions have been established to coordinate economic activity levels at the local level with macroeconomic policy goals at the central level. The article investigates how these institutions were established, and pays special attention to the precise nature of these institutions since they show interesting differences among the otherwise very similar Scandinavian countries. The article pursues the thesis that institutional legacies in intergovernmental relations shape institutions like these, unless decision makers face an institutional tabula rasa. Alternative explanations are briefly considered, but the article concludes by supporting the historical approach to politics.

Introduction

In Scandinavia, local governments are heavyweights in the public economy. They are responsible for a considerably larger share of public expenditure and constitute a considerably larger part of GDP than local governments in most Western countries. Moreover, because of local income tax, they have access to more independent sources of finance than local governments in most Western countries.¹

This fact makes macroeconomic management a challenge. Short-term stabilization is rendered difficult if local government activity is not coordinated with the central government's fiscal policy aims. In extreme cases, local governments may pursue pro rather than anticyclical policies because their expenditure and income sources are sensitive to the cyclical swings of the economy. For instance, since they cannot run deficits, local governments may be forced to increase taxation or lower activity levels in bad times and thereby strengthen depressions."" Long-term structural adjustment of the public sector is made difficult when the responsibility for central

---

*Jens Blom-Hansen, Department of Political Science, University of Aarhus, Universitetsparken, DK-8000 Aarhus C, Denmark.

elements of the public sector is spread out among a large number of entities
beyond the immediate control of the central government.

However, the Scandinavian countries seem to have succeeded in radically decentralizing their public sectors without losing the ability to control macroeconomic performance. The remarkable restoration of public finances in Sweden after the crisis in the early 1990s testifies to this (OECD 1996). So does the ability to keep total public consumption constant in Denmark for a decade after the crisis in public finances in the early 1980s (Christiansen 1990).

The explanation of this puzzle is that institutions to handle the problem are in place. All three Scandinavian countries have established institutions to coordinate the economic activity of local governments with the economic policies of the central government. Intergovernmental corporatism as well as central manipulation of local government income constitute the building blocks of these institutions. In Norway, the local government "revenue system" enables the central government to unilaterally control local government income. In Denmark, coordination of local government expenditure is achieved through annual negotiations between the central government and representatives of local governments, a system known as "budgetary cooperation." In Sweden, coordination of central and local government expenditure is pursued by intergovernmental corporatism during some periods and unilateral central government intervention during other periods. These institutions have made it possible for the Scandinavian countries to radically decentralize their public sectors without losing the ability to control macroeconomic developments.³ How did Sweden, Norway and Denmark succeed in establishing these institutions? This article is devoted to an investigation of that question. It will pay special attention to the precise nature of these institutions since they show some interesting differences among the otherwise very similar Scandinavian countries.

We start by briefly sketching the setting. In all three Scandinavian countries, the decision to establish institutions for macroeconomic control of local governments was taken in the rather special economic-political situation of the 19705. We then move on to a short discussion of how an investigation of these matters can be conducted and what its contents should be. This discussion is followed by a country-by-country analysis. The article then closes with a conclusion.

Background: The 1970s - Economic Policy Making Becomes More Difficult⁴

In the early 19605, public expenditure levels in Scandinavia were still
comparable to those of other European countries. However, at the end of the

decade, the price of the "Scandinavian model" became apparent. Expenditure levels rose well beyond those of other European countries, and kept rising up through the 19705. The trend was not brought to a halt until around 1980. The other side of this welfare boom was rising tax burdens. In the early 19605, Scandinavian tax burdens were not heavier than those found in other European countries. At the end of the decade, tax burdens had outpaced those of other European countries, and taxation kept rising up through the 19705. Around 1980, tax burdens stabilized in Sweden and Norway, but kept rising in Denmark.⁵

This general expansion of the Scandinavian welfare states is relatively well
known. What is probably less known is the extent to which this expansion
took place at the local government level.

In the quarter century after World War 11, all three Scandinavian countries carried out comprehensive local government reforms which fundamentally changed their municipal structures. Before the war, the three countries were divided into innumerable municipalities of highly unequal size, most of which were incapable of carrying out anything but the most basic local tasks. Around 1970, the amalgamation reforms were largely completed. The three countries had now consolidated local governments with sizable populations, stable tax bases and professional staffs (Strömberg & Engen 1996; Rose 1996; Albæk 1996). The restructuring of the local government sectors enabled the Scandinavian central governments to make local authorities the implementing agencies of the welfare state (Kjellberg 1988). Consequently, local governments figured more and more prominently in the public economy in Scandinavia. Their overall economic importance increased dramatically in the late 1960s and up through the 19705, and did not stabilize until around 1980, cf. Figure 1.

Concomitant with the expansion and decentralization of the Scandinavian welfare states was a third phenomenon: a gradual worsening of the macroeconomic situation. The size and form of this problem varied in the three Scandinavian countries, but it was an unmistakable fact. Let us briefly look at the Scandinavian score in this period on four conventional indicators of macroeconomic performance: the deficit on the state budget (measured here as central government debt), the external trade balance, inflation, and unemployment.

Central government debt was relatively stable in all three countries for a long period until the mid-19705. In Denmark, it was even gradually reduced and eventually eliminated up through the 1960s and early 19705. The Scandinavian countries then started running large deficits on their state budgets, and from the mid-1970s to the mid-1980s, central government debt rose dramatically in Sweden and Denmark. In Norway, public debt did not develop into a problem of comparable size, cf. Figure 2a.

The balance of payments deteriorated in all three countries in the early 19705, most dramatically in Norway. In the 1980s, the situation improved in all three countries, again most dramatically in Norway, and in Denmark, things did not improve decisively until the late 1980s, cf. Figure 2b.

All three countries experienced rising inflation from the late 19605, when it
ran in double digits, until the 1980s, when it fell again, cf. Figure 2c.

Finally, unemployment rose dramatically in Denmark in the early 19705,
but was not a serious problem in Sweden and Norway in these years (cf.
Figure 2d).

The expansion of the welfare state and the mounting economic problems were of considerable concern to Scandinavian voters in the 19705. In opinion polls, macroeconomic issues and welfare state questions were consistently given top priority by Scandinavian voters up through the 1970s.⁶ The saliency of these issues also manifested itself in remarkable national election results in Scandinavia in the 19705. They were among the causal factors behind the Danish "earthquake election" in 1973 and the rise of the Danish ultra rightwing Progress Party (Glans 1984; Pedersen 1988), the establishment of a Norwegian ultra right-wing Progress Party in the early 1970s (Bjørklund 1981), and the setbacks of the Swedish Social Democrats in the early 1970s and their final ousting in 1976 after 44 years in power (Petersson 1978). All in

all, we may note that Scandinavian voters gave politicians plenty of incentives
to bring the expanding welfare state and macroeconomic problems
under control.

The general picture that emerges from this sketch of political-economic developments in Scandinavia in the 1970s is one of a rising need for effective macroeconomic management and a decreasing ability to control public activity. At the time, it may have seemed plausible that the Scandinavian countries were set on a course, identifiable in a number of Western countries, towards "welfare backlash" (Wilensky 1975,1976; see also Hibbs & Madsen 1981b) or "political bankruptcy" (Rose & Guy Peters 1978). In retrospect, it is easy to see that these prophecies underestimated the ability of politicians to act. But Scandinavian politicians faced a special challenge, because the welfare state was now a local government phenomenon. Macroeconomic management could no longer be conducted without involving local governments, as they were now responsible for the lion's share of public activity. This may not have been all that clear to the electorate, but it very soon became clear to central policy makers.

Bringing Local Governments into Line - When, by Whom, How, and Why?

Having provided background information on Scandinavian political-economic developments in the 19705, we can now investigate the strategies followed by the Scandinavian central governments to establish institutions to coordinate local government economic activity with macroeconomic policy goals.

The investigation will be based on four questions. First, when did actors at the central level begin to take a macroeconomic interest in overall local government economic activity? Second, who took an interest in this matter - was it limited to a central government concern or were the parliaments involved? Third, how did central actors pursue their interests, i.e., what strategies were chosen to control local government economic activity? Finally, why did central actors make the choices, they did - i.e., how can we understand their choice of strategy?

Before embarking on this investigation, we need to dwell on two problems. The first concerns conceptual clarification. How can we identify and conceptualize a central government strategy to control local government economic activity? Most, if not all, central government activity influences local governments: Laws regulating policy areas administered by local governments (e.g., schools, hospitals, social security) are important to the activity in these areas; general economic policies influence factors such as wage levels and interest rates for local governments too; high central government income tax rates may de facto limit the income tax rates local

governments can set. However, matters such as these influence local government economic activity mainly as a byproduct and will not be treated in this context. We will define central government strategies to control local government economic activity as initiatives which are explicitly, directly, and predominantly aimed at this target. We will distinguish between two types of strategies. On the one hånd, central governments may pursue a coercive strategy, which will typically rely on parliamentary means. On the other hånd, central governments may pursue a cooperative strategy, which will typically rely on some kind of corporatist arrangement in which central governments negotiate with local government representatives.

The second problem concerns explanation of choices. We need to speculate on how we can establish an explanation of strategic choices in intergovernmental relations. As we will see, the answers to the first two questions (dating the choices of strategy and identifying the extent of interest at the central level) are not really surprising. The central governments as well as the parliaments in the three Scandinavian countries began taking interest in macroeconomic control of local governments at about the same time. The picture thus far is one of similarity. Given the relatively similar politicaleconomic situation facing decision makers in the three countries, this is only to be expected. However, the answer to the third question (what strategies were chosen) is troubling. As we will see, Sweden and Denmark chose cooperative strategies, while Norway opted for a coercive strategy. This is puzzling. Why did the three Scandinavian countries, which are so similar in so many respects, choose different ways of handling this problem? We need theoretical guidance to answer this question. The suggested answer here is that institutional legacies shape choices like these, and that only in the case of an institutional tabula rasa will a more actor-oriented approach be necessary.

Let us elaborate the argument. Within New Institutionalism, the fact that political systems are characterized by considerable continuity and inertia and, consequently, that inherited institutional patterns are often important factors in predicting institutional development is now widely recognized (Pierson 1993). However, there is less agreement on why this is so. Different explanations are often caused by different microfoundations. Sociological institutionalism operates with a model of human behavior in which institutions provide prescriptive, evaluative, and obligatory guidelines. In this view, institutional inertia is the result of actors applying existing rules and routines to new situations (March & Olsen 1989, see esp. pp. 34-37). Rational choice institutionalism, on the other hand, operates with homo economicus. In this view, institutions become constraints for self-interest maximizing behavior, and institutional inertia becomes a function of the continuing ability of institutional arrangements to provide advantages to powerful actors

or, alternatively, a function of high transaction costs of change (Christensen
1997; Knight 1992, 173-88; Shepsle 1989).7

Progress in understanding the importance of institutional legacies requires elaboration of the microfoundation of the competing claims. This article takes the position that much can be gained from a view of human behavior which is less extreme and builds on both rational choice and sociological arguments. This position is first and foremost inspired by North and his path dependence argument. In Institutions, Institutional Change and Economic Performance, North argues that institutions define the incentives for different courses of action. Institutions do so by limiting the choices decision makers face. In this way, constraints from the past impose limits on current choices and shape the long-run development of economies and polities (North 1990, 92-105). Again, this explanation of institutional inertia is dependent on the microfoundation. North takes his point of departure in the traditional rational choice behavioral assumptions, but he seriously qualifies them. This issue is set out more systematically in Denzau & North (1994; see also North 1993), which focuses on situations characterized by uncertainty, because "for most of the interesting issues in political and economic markets, uncertainty, not risk, characterizes choice-making" (Denzau & North 1994, 3). The point is that traditional rational choice behavioral assumptions may suffice in situations characterized by a low degree of uncertainty and complexity (such as shopping in a supermarket), but once a "complexity boundary" is crossed, they do not. In situations characterized by high degrees of uncertainty or complexity, the human mind is simply not able to make the deductive rational reasoning requested by traditional rational choice. Instead, the mind relies upon "mental models," i.e., internal representations which the mind creates to interpret the environment. The point then is that historical experiences are important building blocks in mental models. In this way, a community of actors may develop shared mental models that guide their behavior in complex and uncertain situations. Is this historical determinism? No - according to North, "path dependence is a way to narrow conceptually the choice set and link decision making through time. It is not a story of inevitability in which the past neatly predicts the future" (North 1990, 98-99).

Applying this line of thinking to our empirical problem, we may hypothesize the following. Choosing strategies for the first time to coordinate local government overall economic activity with macroeconomic policies is difficult, because these formative moments are complex and uncertain situations. Later choices may be easier because concrete experience is then at hand. When making original choices, decision makers need guiding, and in countries with strong institutional legacies in intergovernmental relations, these legacies will work as guidelines and choices will reflect them. In countries with no institutional legacies in intergovernmental relations, history will not provide any guidelines and the political game will be more open. In

these cases, we will have to apply a more actor-oriented approach and analyze the position taken by each decision-maker in order to explain the choices. As we will argue below, Sweden, Norway and Denmark represent cases at both ends of this continuum.

At this point, we may note that our Scandinavian cases enable us to perform a preliminary test of this argument, since the three Scandinavian countries are similar in so many respects, that a number of variables are held constant. The three countries are probably as "comparable cases" (Lijphart 1971) as we are likely to find in the social sciences when our units of analysis are countries. Of course, not all factors are held constant which is why the test remains preliminary. There is thus an element of hypothesis formulation as well as hypothesis testing in the analysis. However, the comparability of the cases enables us to refute a number of rival explanations advocated in the literature on intergovernmental relations. Let us briefly review these before proceeding with our own hypothesis.

First, patterns of recruitment to national parliaments is a frequently used indicator of central support for local government. The indicator measures the degree to which members of parliaments have served as representatives in local government. The logic behind the argument, rarely explicitly stated, is that overlap between national and local political careers and/or local experience somehow makes members of parliaments sympathetic to local interests (Mackenzie 1954; Knapp 1991). While this logic may be criticized on theoretical grounds,⁸ the recruitment thesis may be rejected in our context solely because local experience was a much more widespread phenomenon in the Norwegian Storting than in the Danish Folketing in the 19705, and had been so for a long time (Skard 1980, Table 11; Damgaard 1977, Table 4.12; Eliassen & Pedersen 1978, 298-300). Consequently, the recruitment thesis cannot explain why Norway chose a strategy that was less sympathetic to local government interests than Denmark.

Second, intra-party relations are often seen as being of decisive importance for intergovernmental relations. One, relatively crude, argument is that relations between political parties' central and local representatives represent a channel of influence which may be used by local councilors and central parliamentarians to influence each other (Gyford 1980; Villadsen 1985). In our context, we may reject this crude version of the party argument because our three countries do not differ in the extent to which political parties are represented at the center and in the localities, i.e., this channel of influence is equally open (or closed) in the three countries. This is easily seen if we compare the share of votes the major Scandinavian parties achieved in national, regional and local elections in the 19705. This is done in Table I.⁹

A more sophisticated version of the intra-party relations argument is to
look at the degree to which political careers at the central level are dependent
on support from local party bases (Page 1991, 108-37; Ashford 1982, 126-69).

While capable of explaining differences in intergovernmental relations between southern and northern countries in Europe, this sophisticated version of the party argument does not seem to be of much help in our context either. As noted by Page (1991, 131-37), the three Scandinavian countries do not really differ in this respect. Therefore, neither the crude nor the sophisticated version of the intra-party relations thesis can explain the variation among our cases.

Third, and finally, the influential policy network approach advocated by Rhodes (1986, see esp. pp. 11-51) does not seem promising in our context either. This approach focuses on the structure of and the actors in the decision making process in intergovernmental relations, and especially on the extent to which local government associations are involved in central decision making. The thesis cannot be directly refuted because of limited evidence and a limited amount of work on this subject in Scandinavia. But, again, the problem appears to be lack of variation among our three cases. In all three Scandinavian countries, the important central government actors are the Finance Ministry and the Interior Ministry, and there are local government organizations which organize all local governments and which to most observers appear influential and relatively integrated in the intergovernmental decision making process.¹⁰ At least in the decisions studied in this

article, the organizations in all three countries were involved and consulted as
a matter of standard operating procedures. Therefore, it seems unlikely that
the policy network approach can explain the variation among our cases.

With these words on conceptual clarification and theoretical guidance,
we can begin the investigation. It will be conducted on a country-by-country
basis.

Sweden

The first major sign in Sweden of emerging central government interest in complete macroeconomic control over local economic activity was the publication of a governmental commission report on macroeconomic stabilization instruments in 1973. The commission devoted a whole chapter to the problem of coordination with local governments. Its point of departure was that "on the background of local governments' large share of the public sector it is, in our view, obvious that better opportunities must be created to coordinate their economic activity with stabilization policies" (SOU 1973:43, 152). In the following years, a number of central government reports were published on local government economy and the problem of macroeconomic coordination. The general impression is one of growing central concern that things are getting out of control. In 1976, a report stated that "continued local government expansion of the same magnitude as in the later part of the 1960s is clearly incompatible with the required balanced development in society in the longer run" (SOU 1976: 45, 23). In 1977, a report sketching scenarios for the macroeconomic implications of continued local government expansion stated that "the perspectives are worrying ... the main impression from all scenarios is that the situation is deteriorating" (SOU 1977:20, 114,123). Later in 1977, a report stressed "the necessity of reducing the expansion of local government expenditure in the coming years" (SOU 1977: 78, 117).

Growing uneasiness with local government economic activity is not only found in the central government, but since the mid-1970s also in the Swedish parliament (Riksdagen). The financial committee in the Riksdag dealt with the question explicitly at least twice in the mid-1970s in connection with its annual treatment of the final state budget (kompletteringspropositionen). In 1976, the financial committee unanimously advocated a significant slowdown of total public consumption, i.e., central as well as local government consumption (FiU 1975/76: 40, see esp. 31-35). One year later, the committee's verdict was even harsher:

Not only private consumption needs to be reduced. The public sector's use of resources cannot be allowed to expand at the same rate as in recent years, in particular local government consumption which in 1975 and 1976 expanded by 4.6 percent each year. For some time, the local government sector has expanded at a faster rate than the central government sector and is now approximately twice as large. Consequently, it is the

committee's opinion that the central government sector alone cannot carry out the necessary
reduction of the public sector's expansion rate. . . . The local government sector's use of
resources must be moderated. Therefore, the committee notes with satisfaction that the
government and local government associations are discussing provisions to reduce
expansion of local government expenditure (FiU 1976/77:30, 54—55).

What strategy did the central government choose to gain macroeconomic
control of local governments? In 1973, a government commission recommended
a corporatist solution:

The most obvious solution is the establishment of a common council by the central government and local governments [i.e., their associations]. Such a council should be responsible for exchange of information, serve as a contact agency for negotiations and function as a forum for common recommendations [concerning economic dispositions by individual local governments] (SOU 1973:43, 168; emphasis in original).

The commission argued that such a council could discuss important macroeconomic initiatives involving local governments in order to gain local government understanding and support for the underlying macroeconomic problems. The commission was of the opinion that it might even be possible to gain positive contributions to macroeconomic management from local governments and to stimulate local governments to consider macroeconomic problems in their own economic dispositions (ibid., 167-68).

The commission did not state its reason for recommending a cooperative strategy, but a few years later, this strategic choice was discussed in detail by another government commission. This commission took as its point of departure that macroeconomic control should focus on the total economic activity by local governments, not on specific policy areas. Therefore, control should take the form of general limits to local government economic activity. The commission discussed two strategies. First, the local government independent taxation right could be legally limited by a tax ceiling. Given the fact that local income tax constitutes Swedish local governments' main income source, this strategy would limit activity by limiting income. The second strategy was to negotiate with local government associations to reach agreements on general taxation and expenditure levels in the local government sector. On the basis of these agreements, the central government and the local government associations could then issue common recommendations to members concerning economic dispositions (SOU 1977:78, 121— 76). We may say that the commission discussed a coercive and a cooperative strategy.

In the end, the commission recommended the cooperative strategy, for the
following reason:

In respect for local self-government, we recommend a solution based on voluntary
agreements between the central government and local government representatives. Legal
action toward the taxation rights of local governments should not be taken (ibid., 176).

The recommendations of the 1977 commission and the 1973 commission were followed. However, a formal council as imagined by the 1973 commission was never established, but informal negotiations between the central government and local government representatives began already in 1973, agreements were reached and common recommendations concerning economic dispositions by local governments were issued in 1973-197'4 and 1976-80 (Murray & Ysander 1983).

How can we explain this basic strategic choice? Why did the Swedes choose cooperative and not coercive means to bring local governments into line? Was the reason as noble as stated by the 1977 commission, i.e., "respect for local government"?

If our theoretical propositions are correct, the answer must be sought in the history of intergovernmental relations in Sweden. Intergovernmental relations in the early 1970s did not represent an institutional tabula rasa. True, as in Norway and Denmark, intergovernmental relations had undergone considerable reform after the war, but the reforms were still relatively modest, at least compared to Denmark's. Further, enactment and implementation of the reforms took several decades (Strömberg & Engen 1996, 241-46). Therefore, history is expected to constitute important guidelines to the original basic choice of strategy in Sweden. But, first a caveat: Is it possible to scour the history of intergovernmental relations for evidence on this matter without this degenerating into an exercise in retrospective rationalization? This is a tricky question which we really cannot answer in the affirmative. To reduce this risk, we will anchor our interpretation of the history of intergovernmental relations in existing scholarly work.

Let us start by looking at the postwar history of the narrow area of intergovernmental economic relations, as this is the area most likely to provide guidelines for the choice of macroeconomic control strategy. The general picture of intergovernmental economic relations in the period 1945— 70 is one of considerable economic autonomy for local governments. This autonomy is based on the independent taxation rights enjoyed by Swedish local governments. Despite steadily increasing expenditure levels and growing economic importance in the public sector, the central government has not tried to curb this right at any point until the 1970s (Hansen 1981; Jonsson 1988). Further, macroeconomic management has traditionally been conducted by the central government largely without involving local governments (Matthiessen 1971). Local autonomy was thus the main feature of intergovernmental economic relations until the 19705, a conclusion also reached by two foreign observers of Swedish intergovernmental affairs in the 1970s (Greenwood 1979; Gold 1977).

However, there are important nuances to this picture. One area of
intergovernmental economic relations has been subject to central control
throughout the postwar period, namely local investment. The instruments

used to gain control have mainly been coercive in nature: quantitative restrictions on local building activity, regulation of credit facilities, earmarked grants, and even occasionally fees on investment activity (Murray 1973; Jonsson 1972, 328-41). Interestingly, however, these coercive means have been supplemented with more cooperative means of control on at least two occasions. In 1960 and again in 1965 corporatist agreements were made between the central government and the local government associations on voluntary restriction of local investment activity (Jonsson 1972, 342^3).

If we include other areas of intergovernmental relations besides the economy, we will see that such corporatist agreements have been used frequently as supplementary means of intergovernmental coordination after the war. Examples include standardization of local government accounting systems in the 1950s (Lönn 1952, 344-46), regulation of public schools in the late 1950s (Landström 1960, 89-90; 1968, 219-20), economic-organizational aspects of the high school reform in 1964 (Landström 1968,218), transferal to counties of public responsibility for care of the mentally handicapped in 1967 (Albinsson 1968, 201-2), and improvement of public child care in the mid-1970s (SOU 1977: 78, 156). In Landstrom's opinion, "cooperation between the central government and the local government associations in the form of negotiations and agreements ... has made it has possible to avoid law making or other kinds of legal regulation" (1976, 163-64). Even if this is so, does this kind of regulation add up to a collective memory which is likely to guide choices in other policy areas? Evidence that this might indeed be the case is given by the 1977 commission (which discussed and finally recommended the establishment of a corporatist arrangement for conducting macroeconomic control of local governments). The commission listed prior examples of intergovernmental agreements and concluded:

It is not unusual to negotiate questions concerning the division of tasks and economic responsibility between the central government and local governments. Through agreements [with the central government], the local government associations are obliged to recommend that local governments follow the guidelines agreed upon in the negotiations. The result is thus likely to be the same as if regulation had been made by legal means (SOU 1977:78, 157; emphasis added).

We may conclude that central government interest in macroeconomic control emerged in the early to the mid-1970s in Sweden and that this interest was located in the central government as well as in the Riksdag. The strategy used to gain macroeconomic control was cooperative, probably because this strategy represented a well-known method of bringing local governments into line.

Norway

Compared to Sweden, macroeconomic control of local governments came
relatively late in Norway. Central interest in the complete economic activity

by local governments emerged quite abruptly in the late 19705. As late as in 1975, the Norwegian central government still expected continued expansion of local government activity and real annual growth rates of 6 percent in local government consumption and investment for the rest of the decade (St. meld, nr. 93/1974-75, 10-15). However, in 1976-78 the macroeconomic situation took a dramatic turn for the worse with rapidly rising deficits on the state budget and the balance of payments (cf. Figure 2). In the opinion of the central government, this meant that the expenditure expansion in the public sector now had to be stopped and that local government activity needed to be brought under central control (Wilhelmsen 1983, 163). In 1978 and 1979, measures were introduced not only against local government investment activity, but also against local government consumption. These latter measures consisted of reductions of local government tax revenue and introduction of obligatory local government contributions to the State Treasury (NOU 1982: 9, 65-72).ⁿ

The measures introduced in 1978 and 1979 were the first clear signals that all local government activity was now of macroeconomic concern to the central government. To some extent, the central government seems to have acted in panic. The steps did not have the intended effects, partly because of the way they were introduced. For instance, the obligatory contributions to the State Treasury were introduced in the middle of the local government budgetary year, which made it difficult for local governments to finance the contributions in the intended way, i.e., by reducing activity. Instead, most local governments financed the contributions through loans and did not reduce activity levels (NOU 1982:9, 61-64).

Although the measures in 1978 and 1979 did not have the intended effects, they certainly placed macroeconomic control of local governments on the political agenda. Macroeconomic control was now linked to another process taking place in these years, namely the efforts to establish a "new income system" for local governments.

Streamlining local government finance into a "new income system" had been announced as a major goal by the central government's commission on reforms in the local government sector already in 1974. Financial streamlining meant transferring the bewildering array of matching grants and reimbursement systems into one general, unconditional block grant to local governments. The reason for this reform was stated as the "financial accountability principle." Matching grants and reimbursement systems were seen as obscuring and dividing the economic responsibility for public tasks and leading to costly and complicated administration. Streamlining would make the public sector more effective and transparent (NOU 1974: 53, 27-35). In 1979, the commission presented its proposal for a "new income system" for the counties. In line with the announcement five years earlier, the commission proposed that existing matching grants and reimbursement

systems be transferred into a general unconditional block grant. The reason stated was the "financial accountability principle" (NOU 1979: 44, 79-82). Three years later, the commission presented its proposal for a "new income system" for the municipalities. The proposal was similar to the proposal concerning the counties, i.e., matching grants and reimbursement systems should be transferred into a general unconditional block grant. Again, the reason given was the "financial accountability principle," but now with an additional attraction: it would facilitate macroeconomic control. The commission now stated that the size of a general, unconditional grant could be manipulated according to the macroeconomic situation:

In the new income system, [the block grant] will replace the earmarked grants . . . and the grants for equalization of tax bases. In the year of introduction, the total size [of the block grant] must be based on the size of the earmarked grants and the equalizing grants in the last year of the old system. The calculated size must then be multiplied by a growth factor which reflects price increases, the general economic situation, and the policy goals of the Storting concerning the economic activity level in the local government sector. In the future, such a growth factor must be used (NOU 1982:15, 143; emphasis added).

The "new income system" could function as a macroeconomic regulator. Local governments would not be able to neutralize reductions in the new block grant by increasing their own taxation, because the central government was already in control of local government tax rates, cf. below.

The commission's proposals were accepted by the government almost intact.¹² The government then introduced the reform to the Norwegian parliament {Stortinget) in two steps. In 1983, the government introduced the general principles of the reform. If these were accepted, the details of the reform would be introduced later. In this introduction of principles, the government stated the "financial accountability principle" as the major reason for the reform, but the implication for macroeconomic control did not go unnoticed:

At the time of introduction, the total size [of the new block grants] will mainly be determined on the basis of the existing grants which are to be incorporated into the new system. The system in its totality must be subject to a general limit for the central government's monetary transfers [to local governments]. This limit is set as part of the general economic policy. Such a system will make it possible to clearly judge the complete monetary transfers from the central government [to local governments] according to fiscal political criteria (St. meld. nr. 26/1983-84, 47)

The government's introduction was treated by the local government committee in the Storting before being discussed in the Storting's plenary sessions. The committee agreed unanimously to the introduction and noted with satisfaction the implications for macroeconomic control:

The committee has noticed that the upper limit to the state's monetary transfers to the local
government sector will now be a budgetary matter subject to macroeconomic goals and
priorities. . . . The committee agrees with this (Innst. S. nr. 305/1983-84, 4).

Six months later Sverre Mauritzen, the speaker of the Storting's local government committee began the general debate in the Storting's plenary session on the government's introduction of principles by stressing the faet that "the committee is unanimous on absolutely every point" (S. tid. 1984-85, 391).

On this background, the enactment of the details of the reforms went smoothly and was agreed to by a unanimous Storting seven months later.¹³ The reform went into effect on January 1, 1986, and since then, the "new income system" has been the cornerstone of the central government's macroeconomic control of local governments. Today, this function of the income system is seen as natural and is almost taken for granted (see e.g. NOU 1996:1, 9).

In sum, the strategy chosen by the central government was to control the
economic activity of local governments by setting limits to local government
income by law. We may say that the chosen strategy was coercive.

How can we understand this choice? A cooperative strategy was never on the agenda. It was never discussed by any of the government commissions, central government itself or by the Storting. Why not? If our theoretical propositions are correct, the answer must be sought in the history of Norwegian intergovernmental relations. Intergovernmental relations in the 1970s did not represent an institutional tabula rasa. True, intergovernmental relations had undergone considerable reform in the postwar period, but as in Sweden, the reforms were still relatively modest. Further, as in Sweden enactment and implementation of the reforms took several decades (Rose 1996, 168-72). Therefore, history is expected to constitute important guidelines to the original basic choice of strategy in Norway.

As in the Swedish case, let us start by looking at the postwar history of the narrow area of intergovernmental economic relations. The general picture of intergovernmental economic relations in the period 1945-70 is one of central regulation. In contrast to Sweden, Norwegian local governments have not enjoyed independent taxation rights, the "cornerstone" of local self-government. As in Sweden, local income tax is traditionally the most important local source of revenue. But, in Norway, local governments do not set tax rates. Since 1911, the central government has set the upper limits to local tax rates, and since the 1950s lower limits have also been imposed. The upper limit has increasingly constrained local taxation as more and more local authorities have found it necessary to use the maximum rate. In 1970, only five local authorities were below the ceiling, and since 1978, all local authorities have used the maximum rate (Hansen 1987; NOU 1997:8, 22-24). Local income tax and central government grants constitute 80-90 percent of local government revenue, so the central government has considerable control over local government finance. In the words of Fevolden & Sørensen (1987, 45), Norway has "a highly revenue-constrained

local government." From an American point of view, Gold concluded that local government "independence seems to be infringed upon much more in Norway than in the other two [Scandinavian] countries" (Gold 1977, 482; see also Kjellberg 1981; and Hansen 1988 on these points). The fact that Norwegian local governments have not experienced the economic autonomy enjoyed by their Swedish counterparts, does not mean that they have been subject to central control on macroeconomic grounds. Despite steadily increasing local expenditure levels and growing local importance in the public sector since World War 11, macroeconomic management has traditionally (i.e., until the 19705), as in Sweden, been conducted by the central government largely without involving local governments (Wilhelmsen 1983).

However, there are important nuances to this picture. One area of intergovernmental economic relations has been subject to central control throughout the postwar period, namely, as in Sweden, local investment. Investment control has always relied on coercive means such as credit quotas in the state banks (which traditionally have provided capital to local governments for investment purposes), regulation of the bond market, and the control by the prefect (fylkesmanneri) over loans raised by individual municipalities (Rattsø 1989). The Norwegian central government has never engaged in corporatist negotiations with local government associations on local investment activity.

So, at least as far as intergovernmental economic relations are concerned, the Norwegian institutional legacy is one of central regulation. To the extent that macroeconomic control has been pursued, it has relied on coercive means. Corporatist means have never been tried because, as noted by Wilhelmsen (1983,165), "the central government has such great control over the factors determining local government income that the need for coordinating arrangements between the central and local governments has been limited." If we look at other areas of intergovernmental relations, the conclusion is the same. Although Norway is one of the classical corporatist countries (cf. Rokkan 1966), there is no record of corporatism in intergovernmental relations.

We may conclude that central interest in macroeconomic control of local governments emerged in Norway in the late 1970s and that interest was located in the central government as well as in the Storting. The strategy used to gain macroeconomic control was coercive, probably because this was the traditional way of bringing local governments into line.

Denmark

In Denmark, the first major signs of central government interest in overall
macroeconomic control of local governments came in 1971. In March, the

government published a comprehensive report sketching long-term macroeconomic
scenarios. One of the conclusions was that:

Local governments are responsible for such a large share of public expenditure - and the share is growing - that macroeconomic management comprising only central government expenditure is grossly incomplete. Without losing the important values of local selfgovernment, ways must be found to incorporate local government budgeting in limiting the development of public sector activities adjusted to macroeconomic requirements (PPI/1971, 36).

Later in 1971, a government commission working with the central government's departmental organization devoted a whole chapter of its report to the problem of coordinating central and local government activity. The commission's point of departure was this."

... the fact that activity levels are rising faster within the policy areas, which are or will be decentralized to local governments, than within the areas administered by the central government implies obvious problems for the central government's and the Folketing's ability to conduct economic policy. ... In the coming years, intensive coordination of the central government's and local governments' economy will be increasingly necessary (Betænkning 629/1971, 36-37).

In the following years, more government reports voiced macroeconomic concern about local government expenditure. In 1973, a government report sketching long-term macroeconomic scenarios was published. This report stated that "... the strongest pressure for the expansion of public activity will be found within the policy areas administered by local governments" and, as a consequence, it saw "an irrefutable need for macrocontrol of the total public activity level" (PPII/1973, 604-5). In 1975, another government commission stated that "because of the far more important role local governments will be playing in the public economy, central planning of local government expansion has become a necessity" (Betænkning 743/1975, 63).

Also the Danish parliament {Folketinget) expressed growing uneasiness with local government economic activity. The first major debate on macroeconomic control of local governments took place in the Folketing already in 1971. The occasion was the government's proposal to introduce an income tax ceiling for local governments. The bourgeois government argued that the tax ceiling was necessary to reduce "the strong growth in local government expenditure" and "growth in the whole public sector's use of resources" (Forslag til lov om den kommunale og amtskommunale beskatning for regnskabsåret 1972-73, column 3729). The major opposition party, the Social Democrats, agreed with the government's goal. Their speaker, Ove Hansen, stated in the Folketing:

My party recognizes that taxation has reached such a level that the situation would be serious to many taxpayers should further tax increases occur. Therefore, we would like to contribute to reducing taxation ... we recognize that intervention is now necessary (Folketingets forhandlinger 1970-71. columns 7975-7978).

The Socialist People's Party did not agree with the specific proposal, but was sympathetic to the overall goal of macroeconomic control. The party's speaker, Sigurd Ømann, stated that "I do not like it [the government's proposal] .. . but maybe this kind of intervention will lead to coordination [of central and local government economic activity] which we must recognize has become necessary" (ibid., column 7985). Only the ultra-leftist party, the Left-Socialists, was outright against the proposal. In the end, the tax ceiling was enacted with the votes from the government and the Social Democrats. Altogether, these parties represented 160 of the 175 members of the Folketing. The Socialist Peoples Party's 11 MPs abstained. Only the Left voted against the tax ceiling.¹⁴

Thus, central interest in gaining macroeconomic control of local governments emerged in Denmark very early in the 19705, in the central government as well as the Folketing. What strategy was chosen to gain control? The introduction of the tax ceiling in 1971 might indicate a coercive strategy, but this is not the case. Denmark never again resorted to such drastic means of macroeconomic control.¹⁵ The 1971 tax ceiling is not easily explained, but must be seen against the rapid expansion of public expenditure and the rapidly rising tax burdens during these years. From 1965-71, Denmark changed from a low-tax country to having the highest tax burden in Europe (PPII/1973, 392). Like Norway in 1978-79, Denmark seems to have panicked in 1971. Neither the government nor the Folketing liked the tax ceiling, but they felt forced to do something until a permanent solution could be found, cf. below.

As in Sweden, the permanent solution was a corporatist arrangement in which the central government and the local government associations would negotiate economic activity levels in the local government sector. Such an arrangement was soon established, and every year since 1972, the government has negotiated economic activity levels in the local government sector with the associations and reached agreements which have been issued as recommendations for economic dispositions to the individual local governments. In the beginning the arrangement only comprised investment activity; from 1975 taxation was included; and from 1979 all local government activity (income and expenditure) was included.¹⁶ As in Sweden, the arrangement was not formalized but consisted of informal negotiations between the central government and local government representatives. We may say that the Danish central government chose a cooperative strategy.

How can we understand this choice? We will argue that Denmark represents a case approximating an institutional tabula rasa in intergovernmental relations in the early 19705. Consequently, historical experiences are largely irrelevant. We may briefly note that, as in Sweden and Norway, macroeconomic management in Denmark has traditionally (i.e., until the 19705) been conducted by the central government largely without involving

local governments (Kjærsgaard & Hansen 1986, 22-^2; Lund & Marker 1978,46-61). Further, as in Sweden and Norway, local investment has been of concern to the Danish central government since the war, and investment control has relied on coercive means (Asmussen 1967; Ingvartsen 1969).

However, because of the comprehensive local government reform taking place in Denmark in the late 19605, intergovernmental relations started with a clean slate in the 19705. As noted by Dente & Kjellberg (1988, 8-14), the Danish local government reform is one of the most comprehensive in the Western world. It was a "global" reform because it covered all aspects of intergovernmental relations (structure, functions, procedures, and financing) as well as all internal matters in local governments (organization, decision making, budgeting). Further, in contrast to Sweden and Norway, the reform in Denmark was a swift process that started in the 1960s and was largely completed by the early 1970s.¹⁷

If our theoretical propositions are correct, history does not provide much of a guideline in such a situation. The political game will be more open and we need to apply an actor-oriented approach to understand the Danish basic strategic choice of how to conduct macroeconomic control of local governments. In the following, we will analyze the position taken by the central government, the main opposition parties in the Folketing, and the local government associations.

Let us start by looking at these actors' position in relation to the tax ceiling
in 1971. The bourgeois government introduced the tax ceiling to the Folketing
with the following words:

. . . cooperation [between the centTal government and the local government associations]
concerning investment budgeting and loan financing must be established. . . . However, this
cannot be done for the local budgets already in 1972-73 . . . therefore, the government has
found it necessary, as a provisional arrangement, to introduce a local tax ceiling for the
budget year 1972-73 (Folketingets forhandlinger 1970-71, column 7918).

This was a firm position. During the whole process, the government kept stressing that the corporatist arrangement was the permanent solution, and that the tax ceiling was a temporary provision necessary only for the next year.^ls

The main opposition party, i.e., the Social Democratic Party, endorsed the proposal, but kept stressing that the temporary character of the tax ceiling was a precondition for their support.¹⁹ Their speaker, Ove Hansen, argued in the Folketing:

. . . he [the Minister of the Interior] has promised that this is a one-off affair. He has promised that a cooperative arrangement between the central government and local governments will be established in the future. We are satisfied with this because we have been in want of such a cooperative arrangement (ibid., column 8079).

The most important Danish local government association (Kommunernes
Landsforening) accepted the tax ceiling, but stated:

However, it has been of decisive importance for the association's accept that it is clearly stated [in the bill] that no legal means to limit the municipalities' taxation will be brought into use after 1972-73 ... a cooperative arrangement [between the central government and local governments] concerning investment budgeting and municipal loan financing must be established (Kommunernes Landsforening 1971, column 3423-3424).

Further, we may note that before the tax ceiling bill was introduced to the Folketing, the government and the local government associations (Kommunernes Landsforening and Amtsrådsforeningen) had negotiated and agreed that a cooperative arrangement should be established. The agreement, which was an official document, stated:

The parties agree to establish a cooperative arrangement concerning investment budgeting
and loan financing. Proposals for procedures for such a cooperative arrangement must
quickly be made (Danish Ministry of the Interior 1971; emphasis in original).

Apparently, the 1971 tax ceiling and the subsequent corporatist arrangement were based on broad consensus. However, it is impossible not to have some doubts about the position of the local government associations. Did the associations really agree with the idea of macroeconomic control of local governments? Even if pursued by cooperative means, macroeconomic control would surely diminish local government autonomy. The limited evidence suggests that the associations felt that they were facing a forced choice and accepted the cooperative strategy as the lesser of two evils. In 1974, Kommunernes Landsforening published a report with a detailed discussion of macroeconomic control of local governments. In this report, the association accepted the overall idea that local government activity may have macroeconomic implications, and it sketched what it saw as the proper way of handling this problem. The solution, "the adjusted system," was to clarify the division of tasks in the public sector, transfer state grants into a general block grant, reduce central regulation of the contents of local government activity, improve local government budgeting procedures, and increase the information about local government activity available to the central government (Kommunernes Landsforening 1974, 25-26). The strategy behind this "adjusted system" was to increase the transparency of the local government sector to improve the central government's opportunity to react to developments at the local level. The implicit consequence was that the necessary macroeconomic adjustments would then only need to involve the central government. Local government autonomy would not have to be violated. According to the association, any other way of dealing with the problem would mean that "little would be left of the decentralization idea on which the municipal reform [in 1970] was based" (ibid., 25). This evidence

suggests that the association was against the whole idea of raacroeconomic
control, and, by implication, especially against coercive means of control.

However, the central government was firmly committed to its course. The indications are that the government was serious when it suggested establishing a corporatist arrangement for exerting macroeconomic control. Apart from the statements made during the Folketing's treatment of the tax ceiling bill (see note 18), the government also stated its intention in official reports. Betænkning 629/1971 advocated "the establishment of permanent procedures for negotiations between the central government and local governments on budgeting" (Betænkning 629/1971, 37). PPI/1971 was less clear, but envisioned "common organs [between the central government and local governments] for conducting expenditure policy" (PPI/1971, 34). A final interesting indication is a speech on Danish economic policy made by Poul Nyboe Andersen, Minister of the Economy in the 1968-1971 bourgeois government, to a meeting of Scandinavian economists in 1972. Here, Nyboe Andersen refers to the corporatist arrangement as the natural way to involve local governments in economic policy making in Denmark (Nyboe Andersen 1972, 135).

Later governments stuck to the strategy. PPII/1973, which was made under social democratic governments, concluded that "steps must be taken as soon as possible to establish a proper budgetary cooperation between the central government and local governments" (PPII/1973, 605; emphasis in original). Betænkning 743/1975, which was made under social democratic as well as liberal governments, stated the necessity of establishing "a negotiating organ where ministers can meet political representatives of local governments" with the purpose of "limiting the total utilization of resources by the public sector" (Betænkning 743/1975, 65). We may say that there was broad agreement at the central level that the strategy to gain macroeconomic control of local governments should be cooperative and not coercive, despite the 1971 tax ceiling.

From the start, it was envisioned that the contents of the corporatist arrangement should be limited to investment activity, cf. agreement between government and local government associations in 1971 (see the quotation above). However, the government soon found that all local government activity needed to be brought under macroeconomic control. PPII/1973 stated: ". . . it is important that budget coordination [between the central government and local governments] is not limited to investment activity, but comprises the whole budget" (PPII/1973, 604; emphasis in original).

On the basis of the available evidence, then, we may conclude that the 1971 government and later governments seriously wanted macroeconomic control of local governments. The local government associations seem to have been opposed to the idea, but accepted cooperative means of control as the lesser of two evils.

All in all, we may conclude that interest in macroeconomic Control emerged at the central level in Denmark in the very early 19705, and that the central government as well as the Folketing were interested. As in Sweden, but in contrast to Norway, the strategy to gain macroeconomic control was cooperative. It is plausible that the reason for this choice of strategy cannot be found in institutional legacies as in Sweden and Norway since the Danish local government reform in the late 1960s radically changed the fabric of intergovernmental relations. Instead, the reason must be sought in the specific positions taken by the main decision making actors.

Conclusion

The three Scandinavian countries have been able to radically decentralize their public sectors without losing the ability to control macroeconomic performance through public income and expenditure. The explanation is that, in all three countries, institutions have been established to coordinate economic activity levels at the local level with macroeconomic policy goals at the central level. The article has investigated how these institutions were established, with special attention to the precise nature of these institutions, since they show interesting differences among the otherwise very similar Scandinavian countries.

The article has demonstrated that the central governments and parliaments in the three Scandinavian countries began taking an interest in macroeconomic control of local government at about the same time. However, the three countries differed in their choice of strategy. While Sweden and Denmark chose cooperative strategies, Norway opted for a coercive strategy. The article argues that institutional legacies in intergovernmental relations shaped these choices in Sweden and Norway. In Denmark, the comprehensive local government reform completely changed the fabric of intergovernmental relations. Denmark thus approximates the probably rare case of an institutional tabula rasa in intergovernmental relations. In such a situation, the political game is more open and a more actor-oriented approach is necessary to explain policy choices.

All in all, the article gives credence to the historical approach to politics (cf. Thelen & Steinmo 1992). In particular, the article supports the claim that in order to understand the institutions we have today, we need to study their origin (cf. Rothstein 1992; Nørgaard 1997). However, we argue that history matters even in formative moments because decision makers are influenced by institutional legacies from related policy fields. Only in the rare case of an institutional tabula rasa are historical experiences irrelevant.

According to Rothstein (1992), the importance of studying institutional
origins is that "in these situations . . . political actors cannot only play the

game, but can also change the rules" (ibid., 174) and that "choosing institutions in a formative moment means that certain political agents are able to structure the future parameters of the political game" (ibid., 190). Our three cases may shed some light on the validity of this proposition. The original institutional choices made in Norway and Denmark of how to conduct macroeconomic control of local governments have proven robust. The institutions established then are now integrated elements of the political systems in these countries. However, the Swedish choice did not stand the test of time. In the early 1980s, the Swedes changed their strategy and have relied on coercive means of macroeconomic control since then. Seen in an historical-institutional perspective, this reversal of an original policy choice is interesting. It seems that there are limits to path dependence. The Swedish strategic reorientation must, however, be the subject of future studies.

---

NOTES

1. For an introduction to the economic importance of Scandinavian local governments see Blom-Hansen (forthcoming), Oulasvirta (1993), and Page (1991, 13-41).

2. For the classic account of this problem, see Hansen & Perloff (1944, chap. 4).

3. For an introduction to macroeconomic control of local governments in Sweden, see Murray (1985) and SOU 1996: 129, esp. chapter 4. For Norway, see Wilhelmsen (1983) and NOU 1997: 8. Vox Denmark, see Blom-Hansen (1996). For an introduction covering all three countries, see Oulasvirta (1993).

4. This heading is inspired by Ruin's thoughtful article on Sweden in the 1970s (Ruin 1982).

5. For a comprehensive and compact survey of these expenditure and revenue trends, see Lane, McKay & Newton (1991, section 5, especially Tables 5.2 and 5.4).

6. For opinion polls of policy priorities of the Scandinavian electorate in the 19705, see the following: Sweden: Bennulf (1991, 42) surveys all opinion polls in the 19705: see also Särlvik (1977). Norway: 1973: Valen & Martinussen (1977. 55); 1977 and 1981: Valen & Aardal (1983, 43); 1985: Aardal & Valen (1989, 44). Denmark: 1971, 1973 and 1975: Sauerberg (1976, 227); 1979: Lindrup & Pedersen (1983, 30). It is important to note that these opinion polls show that macroeconomic issues and welfare state questions were given top priority by the electorate, but not exclusively. Other issues were also considered important, e.g., nuclear power in Sweden and abortion in Norway. This fact probably helps explain why attempts to directly link macroeconomic performance and voter attitudes in Scandinavia in the 1970s have generally failed. For such frustrated attempts, see Madsen (1980); Jonung & Wadcnsjö (1979): Hibbs & Madsen (1981a); Sørensen (1987); Paldam & Schneider (1980).

7. For an introduction to the variants of New Institutionalism, see Hall & Taylor (1996). For an introduction to the competing microfoundations within New Institutionalism, sec Nørgaard (1996).

8. See Page (1991, 56-61). See also Matthews (1985, 24-26).

---

9. The table includes the three largest Scandinavian political parties in the 1970s (in terms of representation at the central level). In all three countries, this happens to be the Social Democratic, the Conservative and the Agrarian-Liberal parties. The percentages in the table indicate the (unvveighted) average of these parties' share of votes in the elections in the 19705. In Sweden (where national, regional and local elections take place at the same time), the elections included are those in 1970, 1973, 1976 and 1979. In Norway, the elections included are the national elections in 1969, 1973 and 1977, the regional elections in 1975 and 1979, and the local elections in 1971, 1975 and 1979. In Denmark, the elections included are the national elections in 1968, 1971, 1973, 1975, 1977 and 1979, and the regional and local elections in 1970, 1974 and 1978.

10. See Page (1991, 45-55) for one such observation.

11. Local government tax revenue vvas reduced by reducing the maximum local government income tax rate. The obligatory contributions to the State Treasury were implemented as contributions to the central governments' income transfer fund (Folketrygden).

12. The only major difference was that the central government proposed that the new block grant be divided into one general block grant and a number of broad sectoral block grants (four broad sectoral block grants for the municipalities and three for the counties). See St. meld. nr. 26/1983-84.

13. On a few points in the large reform complex, the right-wing Progress Party's two members of the Storting dissented. See the parliamentary debate in O.tid. 1984—85, 601^7.

14. In addition to the MPs from the Left-Socialists, one member of the Social Democratic Party, Erhard Jacobsen, broke the party discipline and voted against the proposed tax ceiling.

15. Only once since 1971 has the government imposed a tax ceiling on local governments by law. That was in 1988 and it only comprised the counties.

16. See Nørregaard (1983, appendix 1) for a list of the recommendations issued to local governments on the basis of the agreements in the 19705.

17. For a thorough treatment of the Danish local government reform, see 8011, Ingvartsen, Mikkelsen & Strøm (1991).

18. See the tax ceiling bill and its accompanying explanation (bemærkninger): Forslag til lov om den kommunale og amtskommunale beskatning for regnskabslret 1972-73 as published in Folketingstidende 1970-71, tillæg A, columns 3727-3726. See the Minister of the Interior's introduction of the bill to the Folketing: Folketingstidende 1970-71. Folketingets forhandlinger, columns 7918-7921. See the Minister of the Interior's speech during the Folketing's first treatment of the bill: Folketingstidende 1970-71, Folketingets forhandlinger, columns 7986-7988. See the Minister of the Interior's note to the Folketing's commission treating the bill (Danish Minister of the Interior 1971). See speech by the speaker from the government's Liberal Party (Venstre), Knud Enggaard, during the Folketing's second treatment of the bill: Folketingstidende 1970-71, Folketingets forhandlinger, especially column 8122.

19. See speech by the Social Democratic Party's speaker, Ove Hansen, during the Folketing's first treatment of the bill (Folketingstidende 1970-71, Folketingets forhandlinger, columns 7975-7978). See same speaker's speech during the Folketing's second treatment of the bill (ibid., columns 8078-8080).

ACKNOWLEDGMENTS

For helpful comments on earlier versions, I am grateful to Jørgen Grønnegård Christensen, Peter Nannestad, Asbjørn Sonne Nørgaard, Thomas Pallesen, Lars Dahl Pedersen, Jens Bejer Damgaard, and Carsten Daugbjerg. I am further indebted to two anonymous referees. The research was made possible by a grant from the Danish Social Science Research Council.

REFERENCES

Aardal, B. & Valen, H. 1989. Vclgere, partier og politisk avstand. Oslo: Statistisk Scntralbyrå.

Albæk, E. 1996. "Denmark," in Albæk, E., Rose, L., Strömberg L. & Ståhlberg K., eds.,
Nordie Local Government. Helsinki: The Association of Finnish Local Authorities.

Ashford, D. E. 1982. Britisk Dogmatism and French Pragmatism: Central-Local Policymaking
in the Welfare State. London: George Allen & Unwin.

Asmussen, O. 1967. "Tanker om funktions- og udgiftsdeling mellem stat og kommune,"
Økonomi og Politik 41, 81-95.

Bennulf, M. 1991. "Väljarnas viktigaste frågor 1968-1990," in Holmberg, S. & Weibull, L.,
eds., Politiska opinioner. SOM-rapport 6. Gothenburg: Statsvetenskapliga institutionen.

Betænkning 629/1971. Redegørelse fra arbejdsgruppen vedrørende centraladministrationens
organization. Copenhagen.

Betænkning 743/1975. Planlægning i centraladministrationen. Copenhagen.

Bjørklund, T. 1981. "Høyrebølgens reneste vann," in Bjørklund, T. & Hagtvet, 8., eds.,
Høyrebølgen - epokeskifte i norsk politikk?. Oslo: H. Aschehough & Co.

Blom-Hansen, J. 1996. "Kan budgetsamarbejdet mellem staten og kommunerne forbedres?,"
Nordisk Administrativt Tidsskrift 77, 285-305.

Blom-Hansen, J. Forthcoming. "Avoiding 'The Joint-Decision Trap': Lessons from
Intergovernmental Relations in Scandinavia," European Journal of Political Research.

8011, C, Ingvartsen, 0., Mikkelsen, P. & Strøm, H. 1991. Kommunalreformen i Danmark.
Copenhagen: Kommuneinformation.

Christensen, J. G. 1997. "Interpreting Administrative Change: Bureaucratic Self-interest and
Institutional Inheritance in Government," Governance 10, 143-74.

Christiansen, P. M. 1990. "Udgiftspolitikken i 1980'erne: Fra asymmetri til asymmetrisk
tilpasning?," Politico 22, 442-56.

Damgaard, E. 1977. Folketinget under forandring. Copenhagen: Samfundsvidenskabeligt
Forlag.

Danish Minister of the Interior. 1971. Note to Folketingsudvalget vedrørende lovforslaget om den kommunale og amtskommunale beskatning for regnskabsåret 1972-73. Published as appendix 10 to "Betænkning over forslag til lov om udskrivningsprocenten for kalenderåret 1972, forslag til lov om ændring af lov om sømandsskat og forslag til lov om den kommunale og amtskommunale beskatning for regnskabsåret 1972-73." Folketingstidende 1970-71, tillæg B, columns 3427-3430.

Danish Ministry of the Interior. 1971. "Resultat af forhandlingerne den 6. juli 1971 mellem regeringen og de kommunale organisationer." Published as answer no. 18 from the government to Folketingsudvalget vedrørende lovforslaget om den kommunale og amtskommunale beskatning for regnskabsåret 1972-73. This answer is published in appendix 7 to "Betænkning over forslag til lov om udskrivningsprocenten for kalenderåret 1972, forslag til lov om ændring af lov om sømandsskat og forslag til lov om den kommunale og amtskommunale beskatning for regnskabsåret 1972-73," Folketingstidende 1970-71, tillæg B, columns 3411-3418.

Danmarks Statistik. 1982. Statistiske Efterretninger, nr. A 8: Den offentlige sektors udgifter og
indtægter. Copenhagen.

Danmarks Statistik. 1986. Statistiske Efterretninger: Nationalregnskab, offentlige finanser og
betalingsbalance 1986: 9. Copenhagen.

Danmarks Statistik. 1991. Statistiske Efterretninger: Nationalregnskab, offentlige finanser og
betalingsbalance 1991: 25. Copenhagen.

Danmarks Statistik. Var. years a. Statistisk Tiårsoversigt. Copenhagen.

Dente, B. & Kjellberg, F. 1988. "Introduction," in Dente, B. & Kjellberg. F.. eds.. The
Dynamics of Institutional Change: Local Government Reorganization in Western Democracies.
London: Sage.

Denzau, A. T. & North, D. C. 1994. "Shared Mental Models: Ideologies and Institutions,"
Kyklos 47, 3-31.

Eliassen, K. A. & Pedersen, M. N. 1978. "Professionalization of Legislatures: Long-Term
Change in Political Recruitment in Denmark and Norway," Comparative Studies in Society
and History 20, 286-318.

Fevolden, T. & Sørensen, R. 1987. "Norway," in Page, E. C. & Goldsmith, M. J., eds.. Central
and Local Government Relations: A Comparative Analysis of West European Unitary
States. London: Sage.

FiU 1975/76: 40. Finansutskottets betänkande med anledning av regeringens i propositionen

Folketingets forhandlinger 1970-71. Published in Folketingstidende 1970-71. Copenhagen.

Glans, I. 1984. "Fremskridtspartiet - småborgerlig revolt, högerreaktion eller generell
protest?," in Elklit, J. & Tonsgaard, 0., eds., Valg og vælgeradfærd. Aarhus: Politica.

Gold. S. D. 1977. "Scandinavian Local Income Taxation: Lessons for the United States?,"
Public Finance Quarterly 5, 471-88.

Greenwood, R. 1979. "Relations Between Central and Local Government in Sweden: The
Control of Local Government Expenditure," Public Administration 57, 457-70.

Gyford, J. 1980. "Political Parties and Central-Local Relations," in Jones, G. W., ed., New
Approaches to the Study of Central-Local Government Relationships. Farnborough: Gower.

Hansen, A. H. & Perloff. H. S. 1944. State and Local Finance in the National Economy. New
York: W. W. Norton.

Hansen, T. 1981. "The Dynamics of Local Expenditure Growth: Local Government Finance in
Sweden," in Sharpe, L. J., ed., The Local Fiscal Crisis in Western Europe. London: Sage.

Hansen, T. 1987. "Norwegian Local Tax Policy," Environment and Planning C: Government
and Policy 5, 481-90.

Hansen, T. 1988. "Financial Developments in Norwegian Local Government," in Paddison, R.
& Bailey, S., eds., Local Government Finance. International Perspectives. London:
Routledge.

Hibbs, D. A. & Madsen, H. J. 1981a. "The Impact of Economic Performance on Electoral
Support in Sweden, 1967-1978," Scandinavian Political Studies 4, 33-50.

Hibbs, D. A. & Madsen, H. J. 1981b. "Public Reactions to the Growth of Taxation and
Government Expenditure," World Politics 33, 413-35.

Innst. S. nr. 305/1983-84. Innstilling til Stortinget fra kommunal- og miljøvernkomiteen om et
nyt inntektssystem for kommunene og fylkeskommunene. Published in Stortingsforhandlinger
1983-84. Oslo.

Ingvartsen, O. 1969. "Økonomiske og administrative perspektiver i det nye kommunestyre."
Nationaløkonomisk Tidsskrift 107, 154-82.

Jonnson, E. 1972. Kommunens finanser. Del 1. Stockholm: Ekonomiska Forskningsinstituttet.

Jonsson, E. 1988. "Local Government Finance in Sweden - Present Status and Recent Trends,"
in Paddison, R. & Bailcy, S., eds., Local Government Finance. International Perspectives.
London: Routledge.

Jonung, L. & Wadensjö, E. 1979. "The Effect of Unemployment, Inflation,and Real Income
Growth on Government Popularity in Sweden," Scandinavian Journal of Economics 81,
344_53.

Kjellberg, F. 1981. "The Expansion and Standardization of Local Finance in Norway," in
Sharpe, L. J., ed., The Local Fiscal Crisis in Western Europe. London: Sage.

Kjellberg, F. 1988. "Local Government and the Welfare State: Reorganization in
Scandinavia," in Dente, B. & Kjellberg, F., eds., The Dynamics of Institutional Change.
London: Sage.

Kjærsgaard, K. & Hansen, E. D. 1986. Dansk økonomisk politik. Copenhagen: Nyt Nordisk
Forlag.

Knapp, A. 1991. "The Cumul des mandats, Local Power and Political Parties in France," West
European Politics 14, \S-AQ.

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

Kommunernes Landsforening. 1971. Letter to Folketingsudvalget vedrørende lovforslaget om
den kommunale og amtskommunale beskatning for regnskabsåret 1972-73. Published as

appendix 9 to "Betænkning over forslag til lov om udskrivningsprocenten for kalenderåret 1972, forslag til lov om ændring af lov om sømandsskat og forslag til lov om den kommunale og amtskommunale beskatning for regnskabsåret 1972-73," Folketingstidende 1970-71, tillæg B, columns 3423-3426.

Kommunernes Landsforening. 1974. Kommunerne og samfundsøkonomien. Copenhagen.

Landström, S.-S. 1968. "Kommunförbunden," in Wallin, F., Andersson, H. G. & Andrén, N.,
eds., Kommunerna i förvandling, Stockholm: Almqvist & Wiksell.

Landström, S.-S. 1976. "Staten, kommunförbunden, kommunerne," Nordisk Administrativt
Tidsskrift 57, 161-67.

Lane, J.-E., McKay, D. & Newton, K. 1991. Political Data Handbook. Oxford: Oxford
University Press.

Lijphart, A. 1971. "Comparative Politics and the Comparative Method," American Political
Science Review 65, 682-93.

Lindrup, K. & Pedersen, J. 1983. "Politiske holdninger og partivalg," in Borre, 0., ed.. Efter
vælgerskredet. Aarhus: Politica.

Lund, L. & Marker-Larsen, S. 1978. Finans- og pengepolitik i Danmark 1950-75.
Copenhagen: Nyt Nordisk Forlag.

Lönn, G. 1952. "Svenska landskommunernas forbund," Statsvetenskaplig Tidskrift 55, 331-50.

Madsen, H. J. 1980. "Electoral Outcomes and Macro-Economic Policies: The Scandinavian
Cases," in Whiteley, P., ed., Models of Political Economy. London: Sage.

March, J. G. & Olsen, J. P. 1989. Rediscovering Institutions. New York: The Free Press.

Matthiessen, L. 1971. "Finanspolitikken som stabiliseringspolitiskt instrument." in Lundberg,
E., ed., Svensk finanspolitik i teori och praktik. Stockholm: Aldus/Bonniers.

Murray, R. 1973. "Samordning med kommunerne," in SOU 1973: 45. Budgetreform:
Stabiliseringspolitiska bilagor. Stockholm.

Murray, R. 1985. "Central Control of the Local Government Sector in Sweden," in Gramlich,
E. M. & Ysander, 8.-C, eds., Control of Local Government. Stockholm: The Industrial
Institute for Economic and Social Research.

Murray, R. & Ysander, 8.-C. 1983. "Svenske erfaringer," in Kommunenes rolle i stabiliseringspolitikken.
Nordisk økonomisk forskningsråd. Oslo: Universitetsforlaget.

North, D. C. 1990. Institutions, Institutional Change and Economic Performance, Cambridge:
Cambridge University Press.

North, D. C. 1993. "What Do We Mean by Rationality?," Public Choice 77. 159-62.

NOU 1996:1. Et enklere og mer rettferdig inntektssystem for kommuner og fylkeskommuner.
Oslo.

NOU 1997:8. Om finansiering av kommunesektoren. Oslo.

Nyboe Andersen, P. 1972. "Magtspillet om den økonomiske politik," Statsøkonomisk
Tidsskrift, 129-153.

Nørgaard, A. S. 1996. "Rediscovering Reasonable Rationality in Institutional Analysis."
European Journal of Political Research 29, 31-57.

Nørgaard, A. S. 1997. The Politics of Institutional Control. Ph.D.-Dissertation. Aarhus:
Department of Political Science.

Nørregaard, J. 1983. "Danske erfaringer," in Kommunenes rolle in stabiliseringspolitikken.
Nordisk økonomisk forskningsråd. Oslo: Universitetsforlaget.

O. tid. 1984-85. Odelstingstidende 1984-85. Published in Stortingsforhandlinqer 1984-85.
Oslo.

OECD. 1989. Labour Force Statistics. Paris.

OECD. 1993. Main Economic Indicators: Historical Statistics 1962-1991. Paris.

OECD. 1996. OECD Economic Surveys 1996-97: Sweden. Paris.

Oulasvirta, L. 1993. "Municipal Public Finance in the Nordic Countries," in Gibson, J. &
Batley R., eds., Financing European Local Governments. London: Frank Cass.

Page, E. C. 1991. Localism and Centralism in Europe. Oxford: Oxford University Press.

Paldam, M. & Schneider, F. 1980. "The Macro-Economic Aspects of Government and
Opposition Popularity in Denmark 1957-78," Nationaløkonomisk Tidsskrift 118, 149-70.

Pedersen, M. N. 1988. "The Defeat of All Parties: The Danish Folketing Election, 1973," in
Lawson, K. & Merkl, P. H., eds., When Parties Fail. Princeton, NJ: Princeton University
Press.

Petersson, O. 1978. "The 1976 Election: New Trends in the Swedish Electorate," Scandinavian
Political Studies 1, 109-21.

Pierson, P. 1993. "When Effect Becomes Cause: Policy Feedback and Political Change,"
World Politics 45, 595-628.

PPI/1971. Perspektivplanlægning 1970-1985. Published by the central government. Copenhagen.

PPII/1973. Perspektivplan-redegørelse 1972-1987. Published by the central government.
Copenhagen.

Rattsø, J. 1989. "Kommunene og kredittmarkedet," appendix 1C to NOU 1989:1, Penger og
kreditt i en omstillingstid. Oslo.

Rhodes, R. A. W. 1986. The National World ofLocal Government. London: Allen & Unwin.

Rokkan, S. 1966. "Norway: Numerical Democracy and Corporate Pluralism," in Dahl, R. A.,
ed., Political Oppositions in Western Democracies. New Haven: Yale University Press.

Rose, L. E. 1996. "Norway," in Albæk, E., Rose, L., Strömberg L. & Ståhlberg K., eds., Nordic
Local Government. Helsinki: The Association of Finnish Local Authorities.

Rose, R. & Guy Peters, B. 1978. Can Government Go Bankrupt? New York: Basic Books.

Rothstein, B. 1992. "Explaining Swedish Corporatism: The Formative Moment," Scandinavian
Political Studies 15, 173-91.

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

S. tid. 1984-85. Stortingstidende 1984-85. Published in Stortingsforhandlinger 1984-85. Oslo.

Sauerberg, S. 1976. "Kommunikation til vælgerne - og mellem vælgerne," in Borre, 0.,
Nielsen, H. J., Sauerberg, S. & Worre, T., Vælgerne i 70'erne. Copenhagen: Akademisk
forlag.

Shepsle, K. A. 1989. "Studying Institutions: Some Lessons from the Rational Choice
Approach," Journal of Theoretical Politics 1, 131-47.

Skard, T. 1980. Utvalgt til Stortinget. Oslo: Gyldendal Norsk Forlag.

SOU 1976:45. Kommunernes ekonomi 1960-1972. Stockholm.

SOU 1977:78. Kommunerne: Utbyggnad, utjämning, finansiering. Slutbetänkande av 1976 års
kommunalekonomiska utredning. Stockholm.

SOU 1996:129. Den kommunala självstyrelsen och grundlagen. Stockholm.

St. meld. nr. 93/1974-75. Stortingsmelding nr. 93/1974-75 om en langtidsplan for
kommunenens økonomi 1976-1980. Published in Stortingsforhandlinger 1974-75. Oslo.

St. meld. nr. 26/1983-84. Stortingsmelding nr. 26/1983-84 om et nytt inntektssystem for
kommunene og fylkeskommunene. Published in Stortingsforhandlinger 1983-84. Oslo.

Statistisk Sentralbyrå. 1981. Nasjonalregnskab 1949-1962. B 180. Oslo.

Sentralbyrå.
Statistisk Sentralbyrå. Var. years. Statistisk Årbok. Oslo.

Statistiska 1990.Sektorn:oche. Stockholm.

Centralbyrån. Offentliga Utveckling nuläg
Statistiska Centralbyrån. 1995. Statistisk Årsbok 1996. Stockholm.

Statistiska Centralbyrån. Var. years. Statistisk Årsbok. Stockholm.

Strömberg, L. & Engen, T. 1996. "Sweden," in Albæk, E., Rose, L., Strömberg L. & Ståhlberg
K., eds., Nordic Local Government. Helsinki: The Association of Finnish Local Authorities.

Särlvik, B. 1977. "Recent Electoral Trends in Sweden," in Cerny, K. H., ed., Scandinavia at
the Polls. Washington, DC: American Enterprise Institute.

Sørensen, R. J. 1987. "Macroeconomic Policy and Government Popularity in Norway 1963-1986,"
Scandinavian Political Studies 10, 301-21.

Thelen, K. & Steinmo, S. 1992. "Historical Institutionalism in Comparative Politics," in
Steinmo, S., Thelen, K. & Longstreth, F., eds., Structuring Politics: Historical
Institutionalim in Comparative Analysis. Cambridge: Cambridge University Press.

Valen, H. & Aardal, B. 1983. Et valg i perspektiv. En studie av stortingsvalget 1981. Oslo:
Statistisk Sentralbyrå.

Valen, H. & Martinussen, W. 1977. "Electoral Trends and Foreign Politics in Norway: The
1973 Storting Election and the EEC Issue," in Cerny, K. H., ed., Scandinavia at the Polls.
Washington, DC: American Enterprise Institute.

Villadsen, S. 1985. "Partierne i lokalpolitikken. Statsliggørelse, politisering og lokal
indflydelse," in Villadsen, S., ed., Lokalpolitisk organisering. Copenhagen: Jurist- og
Økonomiforbundets forlag.

Wilensky, H. L. 1975. The Welfare State and Equality. Berkeley. University of California
Press.

Wilensky, H. L. 1976. The "New Corporatism ", Centralization and the Welfare State. London:
Sage.

Wilhelmsen, L. 1983. "Norske erfaringer," in Kommunenes rolle i stabiliseringspolitikken.
Nordisk Økonomisk Forskningsråd. Oslo: Universitetsforlaget.