The Effect of Institutional Arrangements on Political Advocacy

by Swedish and American Nonprofits

Abstract

Using survey data from Swedish and American human service interest groups, this paper explores the impacts of institutional arrangements on advocacy organizations. These two countries were chosen to exemplify corporatist and pluralistic systems.

Results indicate that these different systems of interest group and government relations have significant impacts on how advocacy organizations are structured and operate. As predicted by corporatist and pluralist theory, Swedish groups are different than American groups in many ways relating to their "demographic" attributes, their organizational environment and their relationship with government. In addition, the different institutional arrangements have an impact on what activities Swedish and American groups use to influence policy and the self-perceived effectiveness of those efforts, although not always in the ways expected.

Implications for future research and theory-building are also discussed.

Introduction

One common way to conduct comparative research is to find two cases which are widely divergent on an important variable in order to highlight expected differences. If hypotheses derived from theory hold up in such research, the theory appears sound and can be applied to more ambiguous situations. This paper uses the theories of corporatism and pluralism as exemplified by Sweden and the United States to explore how institutional arrangements affect political advocacy by nonprofit organizations. The results show that these different systems affect many aspects of the way that advocacy organizations are structured and operate.

The paper is divided into several sections. The introduction describes Swedish corporatism and American pluralism and discusses the current situation for voluntary sector advocacy organizations. The paper next details the research methods and general research questions addressed. The third section presents results for human service interest groups in both countries. A final section discusses the results and draws some conclusions about what has been found through this study.

Corporatism and Pluralism

Swedish Corporatism

There is an extensive literature about corporatism in general and Sweden in particular. Most authors promote the view that Sweden is a good example of the phenomenon (Grant, 1985; Lehmbruch, 1984; Schmidt, 1982; Schmitter, 1981; Strich, 1974). Peterson (1977: 59), for example, states: "Corporate structures are particularly well developed in Sweden and their effect on interest group politics is probably more apparent there than in other political systems."

Although the term "corporatism" takes on many meanings in the literature, in the Swedish context we can use Micheletti's formulation:

A minimal definition of corporatism is an integral relationship for decision-making and implementation between the state and encompassing interest organizations. Encompassing interest organizations represent the greater part of the collectivity which they can organize as members. For Sweden, the distinguishing characteristic of corporatism is the incorporation of antagonistic, encompassing interests in state policymaking" (1991:149).

Swedish corporatism is broad in scope and has not changed much in recent years. It has withstood attack from conservative political parties when they controlled parliament in recent years (Bergström, 1991; Hoefer, 1996; Hoefer, 1994; Micheletti, 1991; Rothstein, 1988). While some literature suggests that Sweden is becoming less corporatist (Bergström, 1991; Cawson, 1982, Micheletti, 1991; Peterson & Walker, 1991; Rothstein, 1988), the most recent empirical evidence indicates little change occurring since the mid-1980s (Hoefer, 1996). Organizations, in this system, continue to have official and relatively easy access to decisionmakers (Anton, 1980; Heclo and Madsen, 1987).

Voluntary sector organizations in Sweden are without question part of the corporatist structure. Lundström (1996:139) describes current relations between the voluntary and state sector in terms of "close cooperation and integrative processes" (emphasis in original). In a historical process that accelerated after World War II, the Swedish government took over many voluntary services and made them a part of the welfare state. The voluntary organizations did not oppose this shift in responsibility. (Lundström, 1996). According to Lundström,

Older organizations survived through strategies that include the establishing and developing of activities within certain niches of the field. In addition, more or less new types of organizations were able to establish themselves. These interest organizations and self-help groups often have adopted an ideology related to the labor movement and popular mass movement traditions that were established at the turn of the century (1996:139).

There are three primary corporatist structures in Swedish policy-making. The first are Royal Commissions (utredningar) which the government sets up and funds to study a particular delimited area and to propose legislation. Representatives of interest groups are appointed by the government to be voting members of the commissions. Membership is a highly prized commodity for groups because the basic thrust of policy proposals is shaped here. Groups which are not designated to be members are often allowed to appoint non-voting "experts" to participate in the process. The most similar mechanism in the USA is the use of Presidential commissions, although the Swedish Royal commissions are a much more important and institutionalized part of the legislative process.

The second corporatist structure is the invitation to comment on commission proposals for new policy (the "remiss" system). Open to any group or individual in Sweden, these comments in favor of or against the commission's ideas must be considered by the government and carry substantial weight. In fact, Elvander (1969: 255) found that groups rank the remiss process higher, on average, than being commission members in terms of getting their views across to government. The closest American equivalent is the process of commenting on proposed regulations by government agencies.

The last major corporatist structure is the use of interest group representatives on "lay-person councils" (lekmannastyrelser). These councils are charged with overseeing the implementation of laws and regulations and thus have an important role in day-to-day government operations. They are more powerful than the citizen advisory boards sometimes seen in the United States.

There are currently pressures on the social welfare voluntary sector to change its role in Swedish society. The changes would entail "partly shifting the emphasis of voluntary efforts from being schools for democracy and informers of the people in favor of efforts concerning the production of welfare" (Lundström, 1996:140). These changes will be resisted, according to Lundström, because of the organizations' "grounding in the Swedish tradition of popular mass movements, where basic democracy, member power and, not least, the organization of interest groups are ideals upheld by many" (1996:141).

Helping keep corporatism alive are incentives for established organizations to maintain the current system. Heisler and Kvavik (1974), for example, describe a process of co-optation whereby groups are offered representation in policy-making circles in exchange for supporting the resulting policies, even if the policy is not exactly (or even very close to) what the group desired. In this model, corporatism is fueled by the wishes of the groups to maintain their position within policy-making circles.

While Heisler and Kvavik show why groups wish to maintain the current system, Peterson (1977) stresses the advantages of corporatism to the government. First, the government is able to mold the corporatist environment and governmental structures to improve its bargaining position vis-a-vis interest groups. The government can do this by being willing and able to deny membership on corporatist policy-making bodies to interest groups which do not "act responsibly." Second, Swedish corporatism allows the government to hear a wide range of opinions and uphold the societal injunction to include, in some way, all substantively important interest groups in decision-making relating to a policy area (Peterson, 1977) (for a longer discussion of corporatism in social welfare policymaking in Sweden see Hoefer, 1996 and Hoefer, 1994).

If Lundström is correct, the importance of advocacy efforts in Sweden will continue to overshadow service provision among voluntary organizations in the social welfare sector and the corporatist pattern will endure. This is in line with Heclo and Madsen's conclusion about Swedish policymaking:

There is simply no alternative mechanism to commission work, no alternative structures of corporate representation, no alternative processes for getting heard and getting along in public life (1987:22).

American Pluralism

One of the important functions of nonprofit organizations in the United States has traditionally been advocacy, particularly for increased government funding for services to individuals, although service provision and other roles are also important (Kramer, 1981). In this paper, we focus on the type of American nonprofit advocacy organizations often referred to as interest groups. Interest groups in America have been defined as "any group that is based on one or more shared attitudes and makes certain claims upon other groups or organizations in society (Truman, 1971:33). In order to do this, they "focus mainly on the lawmaking phase of the governmental process" (Hrebenar, 1997:10).

Both the positive and the negative aspects of American interest groups are detailed in the voluminous literature on these organizations. While there is controversy about many elements of interest groups, it is clear that groups are not written into the American constitution, nor are there such well developed official pathways to policy-making power for them in the United States as there are in Sweden. Interest groups in America are important in the policy-making process but their place is outside the formal limits of the Constitution. They have a place in the process only to the extent that other actors allow them to have a place.

In order to be included in the process, interest groups in the United States usually bring with them key resources which they share with decision-makers. These key resources are funds, information and loyal members.

Funds are important to elected officials as it is very expensive to run for office. Campaign contributions from groups are usually welcome and are considered important in gaining access to these officials after the election is won. While perhaps commonly believed by the public, there is no scholarly evidence that legislative votes are "purchased" with campaign funding.

Information is also an important resource for interest groups. Policy-makers want information (either to support a position already held or to aid in taking a position) and an interest group with the appropriate data is in a strong position to influence a decision-maker, either in the legislature or in the bureaucracy.

Loyal members allow interest group leaders to use a variety of tactics. Instead of (or in addition to) presenting information directly to policy-makers, for example, a group could mail the same information to its members and have them contact their own legislators or the proper executive office. Loyal members are also more easily "delivered" as votes to elected officials.

The doctrine of pluralism indicates that there are many groups in the American policy-making process, all working to have their ideas adopted by elected officials. Decision-makers respond to what they see before them and maintain order between the contending groups. All groups thus have some chance of having their ideas adopted and their interests satisfied although groups with more money, information and/or members have better chances.

This concept of a pluralistic political system is not shared by all observers of the American scene, many of whom argue that policies in America are generally made in favor of the wealthy and well-connected. Advocates of this "elite" view maintain that the most important issues are not up to a decision by elected officials. Only a few groups can ever hope to be influential and these are agents of the upper classes.

Whichever is the "truth" about pluralism and elitism at the very highest levels of decision-making, decisions made by Congress and state legislatures, the President, governors and other actors contacted by interest groups are important to many American citizens, especially those receiving benefits from human service programs. Funding can be increased or cut. Benefits can be changed from in-kind to cash. Eligibility can be broadened or narrowed. Tax laws can be altered to include program benefits or exclude them. Thus, the ultimate truth of policy-making in the United States is that elected officials make important choices and interest groups often influence the final outcomes of those choices.

Researchers note several recent changes in the American interest group system. Perhaps the most important is that the number of groups has grown tremendously in recent years (Loomis and Ciglar, 1995; Hrebenar, 1997; Petracca, 1992; Walker, 1983). Within this tremendous increase has also been a shift in the types of groups, with a particular increase in the number of groups representing "the public interest" or interests of persons who have heretofore been more-or-less excluded from the political process (Loomis and Ciglar, 1995; Hrebenar, 1997; Petracca, 1992; Walker, 1983).

No one knows precisely how many interest groups there are in the United States at the local, state and national levels. In 1996, however, it was estimated that there were over 22,000 interest groups at the national level (Hrebenar, 1997). Of these, about 1,900 were classified as being active in social welfare issues, up from nearly 1,000 in 1980 (Hrebenar, 1997: 15).

A second major shift affecting the operations of human service advocacy organizations is that the technology available to groups has changed. Computers allow more targeted communication to organizational members and decision-makers. Once-esoteric communication mechanisms such as fax machines, cellular phones and electronic communication make possible enhanced grass-roots strategies and speedier responses to legislative events (Loomis and Ciglar, 1995; Hrebenar, 1997; Petracca, 1992).

The end result of these (and other) changes is that "while the proliferation of interest groups has resulted in increased representation, it has also resulted in decreased accountability and responsibility" (Hrebenar, 1997: 334). Wilson argues that the American system is a reflection of political culture.

In its interest group system, the United States reflects a balance that might be said to be true of its political system more broadly: emphasis is placed more on representation and less on effective governance. No interest group system exceeds the capacity of the American one to represent effectively a wide diversity of views.In contrast, the opportunities for producing better governance or improved growth through the sort of partnership between government and economic interests found in Japan or Sweden are absent in the United States (1992:95).

Having examined the role of nonprofit advocacy organizations under Swedish corporatism and American pluralism, we now turn to discussing the methods of the current study and specify the research questions addressed.

Methods

This research project employed surveys of human service interest groups in Sweden and the United States. The questionnaires were as identical to each other as possible while reflecting differences in political institutions. The surveys were completed in 1994 by 61 Swedish organizations and 127 American organizations,. This represents a 60% return rate for the Swedish survey and a 40% rate for the American survey.

For Sweden, the primary way that organizations were selected for inclusion in the research was if they had used the remiss system to comment on government human services policy proposals during the 1980s. This list was augmented by selecting a few additional groups listed in the Stockholm phone book under any one of many different headings reflecting advocacy-oriented organizations. American organizations were included if they were listed in the 1992 Washington Information Directory (a list of thousands of interest groups) as being active in social welfare policy and if the group confirmed that it lobbied the executive branch as well as the legislative branch.

For both countries, a typical mail survey process was used in this research: an initial mailing to all groups, a postcard "Thank you"/reminder ten days later, and a second full mailing to all non-respondents two weeks later (Fowler, 1988).

This research is unique in having identical survey information on a broad range of issues regarding advocacy organizations active in social welfare issues from Sweden and the United States. The American survey is similar to that used by Walker (1983); it was adapted to fit the Swedish situation by the author. Before being used in Sweden, however, three Swedish political scientists read it and suggested various improvements.

Research Question

This paper is a first look at the data collected by these surveys. The overarching question we address is: "Do the structures of corporatism and pluralism affect the demographics, environment, activities and effectiveness of human service advocacy groups?"

Results

This section is divided into four areas, focusing on the topics of respondent demographics, group environment, group advocacy activities and self-reported group advocacy effectiveness. Each subsection begins with a description of several hypotheses about the ways that corporatism and pluralism should be reflected in the data. Results are then shown to either support or reject the hypotheses.

Respondent Demographics

Given the nature of a corporatist system in which "encompassing organizations" are granted access to the policymaking processes and the length of time that Sweden's system has been corporatist, we believe that there is a strong incentive for interests to have organized quite a while ago and to have maintained an all-encompassing organization. Groups active in policymaking generally derive their legitimacy from representing (and, in some ways, controlling) citizens who are members. Representation requires members; control is aided by an organization that is not merely a national group, but one that has local sub-units.

A pluralist system, however, should look different. Groups have little incentive to be all-encompassing; groups are more likely to try to exploit narrow niches in the policy space (Browne, 1991). This would lead to a number of small organizations, both in terms of staff and membership (if any; organizations may not have any members). Those organizations that are membership groups have little incentive to establish local or state chapters that might compete for contributions of scarce money and time.

When describing Swedish groups, then, we should expect that they will be older (on average) than US groups, more "member-oriented" and structured with more subunits. We also suspect that the American system will be more dynamic, with larger increases in staff size and membership in recent years.

Table 1 shows data that support these generalizations. The median age for Swedish groups is over twice that of American groups (52.0 years compared to 23.5 years). All of the Swedish groups have members compared to only 74% of US organizations. More of the Swedish organizations have subunits and their median staff size is larger than the median staff size of American groups. (American organizations do have a larger mean staff size, but this is due to two organizations with extremely large staffs.)

Table 1: Demographic characteristics of the sample

Demographic Category	Sweden (n = 61)	USA (n = 127)	P value
Median Age	52.0 years	23.5 years	.000 A
Have Members?	100%	74%	N/A
Mean Number Individual Members	96,276	616,001	.079 B
Mean Number Organizational Members	4,761	10,709	.103 B
% Groups with an Increase in Membership Size in Last 5 years	38%	54%	.000 A
Have Subunits?	82%	52%	.000 A
Median Staff Size	16 FTEs	11.5 FTEs	N/A
Mean Staff Size	49 FTEs	64 FTEs	.035 B
% Groups with an Increase in Staff Size in Last 5 years	25%	49%	.000 A

Notes: A - statistical test is a chi-square test.

B - statistical test is an Eta test.

Complete information on the distributions used to calculate the statistics are available from the author.

The data also indicate that the American system has seen a greater percentage of groups increasing their membership and their staff size in the recent past compared to the Swedish groups. This dynamism may be related to the fact that the Swedish interest group system is already very crowded and the groups have already done a good job of having nearly all willing persons join the organizations they are eligible for.

Group Environment

One of the key components of an interest group's existence is the type of environment within which it operates. We expect that a group in a corporatist environment has a secure niche, that competition with other groups for members and resources is limited and that conflict is limited, as actors in the political system attempt to accommodate their needs with the needs of all. In addition, because corporatism includes groups in the decisionmaking process no matter which political party is in power, we should see little change in group-government relations after a shift in control of parliament.

A different picture emerges for a pluralist system. The image that emerges is that of a very competitive situation, where actors seek to gain as many spoils from the system as they can. New groups begin, hoping to draw members and resources away from established groups by offering a different array of benefits and by claiming to advocate more effectively. Also, because interest groups tend to be linked with one political party or another (Peterson and Walker, 1991), their access will vary according to which one is in control.

The data indicate that these hypotheses are correct. When asked how much competition groups encounter in obtaining members and other resources, US groups indicate that they are in a much more competitive environment than are Swedish groups (chi-square of 18.815, p = .000). Nearly one-third of American groups (32%) say that they are in continuous competition compared to just one-fifth of Swedish groups(20%). Over half of Swedish groups (54%) say that they have little competition; only one-fourth (23%) of American groups respond this way. All other groups indicate that they have some competition. (46% of American groups and 26% of Swedish groups). Table 2 examines the level of conflict in the environment perceived by group leaders.

Table 2: Elements of the Environment

	Sweden (n = 61)	USA (n = 127)	P value
Intense conflict exists	2.33	3.04	.002
Elected officials oppose group	2.82	3.53	.001
Other groups oppose group	2.75	3.51	.001
Government agencies oppose group	2.20	2.65	.022

NOTE: Degree of agreement is measured on a Likert type scale with 1 being "not at all a good description of our group's situation" to 5 being "a very good description of our group's situation." P-value is derived from a T-test.

For each of the four questions indicating the perceived level of conflict and potential sources for conflict, American groups report more conflict in their environment than do their Swedish counterparts. American group leaders are more likely to believe that intense conflict exists regarding the issues of importance to their group. They are also more likely to believe that elected officials, other groups and government agencies oppose them than are Swedish leaders.

As predicted, group-government relations are different in Sweden and the US. Almost all (95%) Swedish groups say that they are contacted by government agencies on a regular basis, compared to about three-fourths (72%) of American groups (chi-square = 14.143, p = .000). Groups report similar levels of cooperation from the agency they have the most contact with (54% of Swedish groups say that that agency "normally cooperates" compared to 62% of American groups), but this may be an artifact of having Democrats in power during the time of the survey. American groups indicate that cooperation is dependent on which party is in power; Swedish groups do not.

When asked how cooperation was affected by the most recent shift in control of the Administration (from Republican George Bush to Democratic Bill Clinton in the USA and from Social Democrats to a conservative coalition in Sweden), two-thirds of American human service interest groups (66%) found an increase in cooperation. Only one-fourth of Swedish groups reported an increase in cooperation. Over half of Swedish groups (53%) , on the other hand, said there was no change in levels of cooperation, compared to one-fourth (24%) of US groups. One-fifth (20%) of Swedish groups did report less cooperation as the liberal party lost power. Only 10% of American groups indicated that cooperation was less under the Democrats than the Republicans (chi-square 34.401, p < .000).

The information in this section points to clear differences in how interest groups see their environment and the ways that they are connected to government agencies. As predicted, Swedish groups feel less competition and less conflict than do American groups. Swedish organizations have greater continuity in their relationships with government agencies than do American groups, even after significant political changes occur.

Group Advocacy Activities

With every issue they confront, interest groups are faced with the decision of how to respond to affect the situation. There are many constraints on their possible reactions. Perhaps the most important one is that group leaders must choose tactics that will maintain and perhaps enhance the flow of resources to the organization.

We expect Swedish corporatism to foster a particular set of tactics because of its emphasis on close working relationships between government and groups, reaching consensus before making proposals public, the parliamentary system of government and of working through the particular Swedish corporatist institutions such as Royal Commissions and the remiss process.

Elements of the pluralistic American system that might lead to a different set of tactics being used include the emphasis on the power of individual legislators, the need to remain officially nonpartisan in order to protect a group's nonprofit status and the importance of litigation for creating legal precedents.

Table 3 indicates the perceived importance of nine tactics that could be used to affect policy. Two of the tactics, "Work on remiss submissions" and "Work on Royal Commissions," are applicable only in Sweden. The other tactics can be employed in both countries.

Table 3: Activities used to affect policy decisions

	Sweden (n = 61)	USA (n = 127)	P value
Work on remiss submissions	5.14	n/a	n/a
Affect public opinion through the mass media	4.82	4.18	.005
Work to affect Royal Commissions	4.62	n/a	n/a
Work with bureaucracy	4.58	4.94	.11
Work with legislators and their staff	4.29	5.17	.000
Develop a policy consensus among experts	3.36	4.01	.010
Litigate in court	2.70	2.20	.051
Protest and demonstrate	2.31	1.66	.001
Elect a particular party to office	1.44	1.56	.544

NOTE: Importance of tactic is measured on a Likert type scale with 0 being "tactic is not used" to 6, "one of the most important tactics." P-value is derived from a T-test.

As before, there are significant differences in group responses to this set of questions. Swedish organizations believe that "affecting public opinion through the use of the mass media," "litigating in court," and "using protests and demonstrations" are all significantly more important ways of influencing policy decisions than do American groups. American groups, on the other hand, are significantly more likely to believe that "working with legislators and their staffs," and "developing a policy consensus among experts" are more useful. Americans also favor "working with the bureaucracy," and "working to elect a particular party" but not at a statistically significant level.

Only one of these differences was in the expected direction. Working with legislators and their staffs is more important in the American system due to single member districts and relatively weak political parties. The Swedish parliamentary system, with strong party allegiance, works against emphasizing lobbying individual representatives.

Most of the differences are at least somewhat surprising. Swedish groups put much stock in using the mass media to affect public opinion, a tactic that does not seem to be a very "corporatist" way to influence policymaking. Perhaps the explanation is that interest groups seek to influence the public as a way of "controlling" them into accepting a policy measure already agreed upon by groups and the government.

It is also surprising that Swedish groups believe that litigating in court is more important than American groups do. America is a much more litigious society than is Sweden and it seems that Congress and the President are relieved when the courts take on some of the more divisive policy issues.

Finally, two tactics relate to consensus building and formal conflict. One would expect that corporatist Sweden would be the country where building a consensus among experts and de-emphasizing protests would be more highly prized, compared to a fractious, pluralist USA. The opposite is the case however. Consensus building is esteemed more highly and protests are valued less in America than in Sweden.

Advocacy Effectiveness

The end result of any interest group action, organizational leaders and members hope, is policy that is in line with the group's position. While interest group effectiveness is difficult to measure, there have been some efforts (for more detail on this topic, see Hoefer, 1994). The approach used in this research is self-reported effectiveness. Despite its apparent shortcomings in terms of objectivity it has the benefits of being easily obtained and a good starting point for comparisons. It also allows people with much insight and little reason to exaggerate on an anonymous survey to provide their input.

Given the literature on corporatism, we might expect groups to be frequently successful in their advocacy efforts. Their positions are arrived at after considerable internal debate and are supposed to take into account the other interests represented in policy-making. This type of situation brings forth an image of self-effacing efficacy. Leaders would ask only for what is reasonable to begin with and thus would increase the chances of getting it.

American pluralism, on the other hand, is extremely conflictual and, with the number of groups growing quickly, we expect American groups to be less successful, on average, than are Swedish organizations.

We note some similarities between Swedish and American groups in terms of how much of their time they spend on trying to influence policy (as opposed to providing member benefits, finding funding for the organization, and so on). Swedish groups report using 40% of the group's time in that way compared to US groups spending 31% in trying to affect policy.

The same percentage of groups in both countries reported that they were making more policy-change efforts compared to five years previously (49%). Similar amounts were reporting making the same number of efforts (46% in Sweden, 44% in the US) and fewer efforts (5% in Sweden, 7% in the US).

Surprisingly, though, Swedish leaders report lower levels of political efficacy than do their American counterparts. Swedes say that they are effective in just 26% of their efforts, whereas Americans say that they are effective 53% of the time (t = 7.178, p < .000).

Discussion and Conclusions

The question this paper addresses, "Are nonprofit advocacy groups affected by institutional arrangements?" can be answered with a simple and emphatic "Yes." As predicted by theory, Swedish groups are different than American groups in terms of demographic characteristics and perceptions of their environment. Specific variables showing a significant difference are age, level of centralization and level of "dynamism" (as shown by lower rates of growth in staff and membership in the five years prior to the survey). Swedish group leaders feel that their environment is less conflictual than do American group leaders. Swedish groups' relationship with government is also significantly less constrained by which party is in power.

The different institutional arrangements have an impact on what activities Swedish and American groups undertake to influence policy, although not always in the ways expected. Corporatism and pluralism have an impact on the self-perceived effectiveness of those advocacy efforts, with American leaders saying that they are more effective than do Swedish leaders.

Most, though not all, of these results are in line with predictions based on a detailed examination of corporatist and pluralist theory. There are several implications of these findings for researchers of advocacy organizations. First, they show the need to understand the context within which interest groups operate in order to understand their actions. Ignoring the larger context is easy enough to do: anyone can collect similar data from different countries. Doing only this may, however, lead to "mindless" research. Comparative research is much more valuable when it has a conceptual grounding.

Second, the particular concepts chosen here, corporatism and pluralism, are shown to have considerable utility in pointing to potential differences in the world of human service advocacy organizations. An ongoing problem with much of the literature on corporatism and pluralism is that testable propositions are not derived and empirical research is not conducted to show whether these concepts are useful (for an exception to this generalization, see Hoefer, 1996 and Hoefer, 1994). As Williamson (1989:184) states, regarding corporatist writings: "the empirical findings, and more importantly, the theoretical and conceptual questions and leads that these can provide, are relatively thin on the ground". This research adds to the current "thin" layer and points the way for other researchers to follow.

Third, despite the successful interpretations of theory, not all of our hypotheses were supported. Whether this requires altering the theory or improving the interpretation is not yet clear. It is clear, however, that more research would be helpful to address these issues.

There are certainly limitations to this study as a test of the effects of corporatism and pluralism on interest group advocacy. The primary one is that we equate corporatism with Sweden and pluralism with the United States. While the literature agrees that these countries are good examples of these approaches to interest group and government relations, there are other countries that could have been chosen to represent the concepts. Additional research should be conducted to extract what is inherent to corporatism and pluralism, and what is unique to Sweden and the United States.

Another limitation is the way that some concepts were operationalized. The questionnaire was originally designed for use in the United States and, though adapted for the Swedish situation, may contain hidden pluralist biases that skew the results.

The type of Third Sector organization studied here, interest groups, is a very specialized type of organization, devoting a considerable amount of the organization's time to advocacy. There are many other organizations that conduct advocacy efforts, though not at such a high priority level. The answers those bodies would provide may or may not be similar to the respondents chosen for this research.

Despite these limitations, the research results described here are among the few available that compare nonprofit advocacy groups in different countries in a way that also tests theory. As such, it may lead to additional, and more complete efforts. Particularly useful would be to use additional countries that have corporatist and pluralist systems and to use a wider variety of nonprofit groups.

References

Anton, T. 1980. Administered politics: Elite political culture in Sweden. Boston: Martin Nijhoff.

Bergström, H. 1991. Sweden's politics and party system at the crossroads. West European politics, 14, 8-30.

Browne, W. 1991. Issue niches and the limits of interest group influence. In Ciglar, A. and Loomis, B., eds. Interest group politics, 3rd ed. Washington, DC: Congressional Quarterly Press, 345-370.

Cawson, A. 1982. Corporatism and welfare: Social policy and state intervention in Britain. London: Heinemann Educational Books.

Elvander, N. 1969. Intresseorganisationer I Dagens Sverige [Interest groups in today's Sweden], 2nd ed. Lund: CWK Gleerup Bokförlag.

Fowler, F. J., Jr. 1988. Survey Research Methods. Newbury Park, CA: Sage.

Grant, W. 1985. The political economy of corporatism. London: MacMillan.

Heclo, H. & Madsen, H. 1987. Policy and politics in Sweden: Principled pragmatism. Philadelphia: Temple University Press.

Heisler, M. & Kvavik, R. 1974. Patterns of European politics: The 'European polity' model. In

Heisler, M., ed. Politics in Europe: Structures and processes in some post-industrial democracies. New York: David McCay Co., 27-89.

Hoefer, R. 1996. Swedish corporatism in social welfare policy, 1986:1994: An empirical examination. Scandinavian political studies, 19(1), 67-80.

Hoefer, R. 1994. Corporatism, pluralism and Swedish interest group influence in social welfare policymaking. Governance, 7(2), 165-181.

Hrebenar, R. 1997. Interest group politics in America, 3rd ed. Armonk, NY: M. E. Sharpe.

Kramer, R. 1981. Voluntary agencies in the welfare state. Berkeley, CA: University of California Press.

Lehmbruch, G. 1984. Concertation and the structure of corporatist networks. In Goldthorpe, J., ed., Order and conflict in contemporary capitalism. Oxford: Clarendon Press.

Loomis, B. & Ciglar, A. 1995. Introduction: The changing nature of interest group politics. In Ciglar, A. & Loomis, B., eds., Interest group politics, 4th ed.. Washington, DC: Congressional Quarterly Press, 1-31.

Lundström, T. 1996. The state and voluntary social work in Sweden. Voluntas, 7(2), 123-146.

Micheletti, M. 1991. Swedish corporatism at a crossroads: The impact of new politics and new social movements. West European politics, 14, 144-165.

Petracca, M. 1992. The rediscovery of interest group politics. In Petracca, M. The politics of interests: Interest groups transformed. Boulder, CO: Westview Press, 3-31.

Peterson, E. 1977. Interest group incorporation in Sweden: A summary of arguments and findings. Paper presented at the Annual Meeting of the American Political Science Association, Washington, DC, September.

Peterson, M. & Walker, J., Jr. 1991. Interest groups and the Reagan presidency. In Walker, J., Jr., Mobilizing interest groups in America: Patrons, professions and social movements. Ann Arbor: University of Michigan Press, 141-156.

Rothstein, B. 1988. State and capital in Sweden: the importance of corporatist arrangements. Scandinavian political studies, 11, 235-260.

Schmidt, M. 1982 Does corporatism matter? Economic crisis, politics and rates of unemployment in capitalist democracies in the 1970s. In Lehmbruch, G. & Schmitter, P. eds. Patterns of corporatist policymaking. Beverly Hills: Sage. 237-258.

Schmitter, P. 1981. Corporatism is dead! Long live corporatism!. Government and opposition, (24)1, 54-73.

Strich, T. 1974. Introduction. In Pike, F. & Strich, T., eds. The New Corporatism. London: University of Notre Dame, xiii-xxii.

Truman, D. 1971. The governmental process. New York: Knopf.

Walker, J., Jr. 1983. The origins and maintenance of interest groups in America. American political science review, 77, 390-406.

Williamson, P. 1989. Corporatism in perspective: An introductory guide to corporatist theory. Newbury Park: Sage.

Wilson, G. 1992. American interest groups in comparative perspective. In Petracca, M. The politics of interests: Interest groups transformed. Boulder, CO: Westview Press, 80-95.