Introduction

Policy scholars have been much attracted to cause and effect explanations of policy change.  Following one of the most common methodology in the social sciences, a theory is constructed about factors postulated to cause changes in policy.  These theories are frequently constructed in comprehensive terms at a fairly high level of generalization regarding application, and therefore can be called general theories of the determinants of policy change.  Each is often formulated so as to appear to provide the whole of the explanation of a policy change. The implication very often is that each is a better candidate than any of the others as an explanation of changes in policy, and thus they are in some sense mutually exclusive.  As we will see, the assumption of mutual exclusivity is not essential in most cases, and the tendency to imply that this is the case is often unfortunate.

This chapter will examine in summary form some of the most commonly advanced theories, and assess their strengths and weakness as general theories of policy change.

Structuralist Theories

Structuralism in policy terms means that change takes place without any agent of the state consciously deciding on the change or being aware of the reasons for the change, and without any group having to actively advocate for the change.  Policy does not have its origins in state agents or in active and conscious group pressure.  Change is not dependent upon the consciousness of individuals or groups of problems or needs, nor is it dependent on action by them.  It is the economic or political structure that determines policy, and these structures can be objectively observed and differentiated. And the structures themselves are not the result of the effort of individuals, groups or agencies.  They are the result of impersonal economic and political forces, beyond the reach of human agency.

Structuralists do not all approach the determination of policy change in the same way, however.  In the following three of the most significant structuralists approaches are summarized.

1.       Marxism

The form of structuralism that has received the greatest attention over the years is Marxism, named after its inventor, Karl Marx.  Marxism draws upon historical materialism as a theory of history, in which the structure of society is based on material conditions within society.  Societies inevitably and necessarily progress through time in stages from feudal to capitalist to communist.  The key distinguishing characteristics of each are the division of society into classes and the organization of ownership of the means of production.  These explain the structures of society, which change not from state or organized decisions but from immutable historical circumstances as progress is made from one stage to another.

The stage that is most interesting for our purposes is capitalism, since capitalism defines an industrialized democratic society such as our own.  Our economic system is essentially capitalist.  The means of production in a capitalist economy are the basic things used to produce things for sale into markets, in pursuit of profit.  The primary means of production are capital and labour.  Capital is saved up labour and is constituted in the form of large concentrations of equipment, technology, and finance in the hands of a single or small concentrations of individuals.  Capital is the property of capitalists who form the capitalist class, and in capitalist societies property is for the exclusive benefit and under the exclusive control of the owners.

Labour is essential to production.  In order to make capitalism work, labour must be commodified.  These means that a person’s capacity to work, as a service, must be subject to purchase and sale.  In organizing production, the capitalist combines capital, which can be owned and labour, the services of which must be purchased as it is needed.  A surplus or profit arises from controlling production through the ownership of capital and from paying labour less than its contribution to the value of what is produced.  Upon selling the product, the capitalist thus realizes a surplus, which is the surplus product of labour effort.  When labour is commodified, the only means it might have of capturing some part of the surplus, which arises from its effort and time applied to capital, is through wages, which is the price of labour.  But this isn’t possible because the capitalist controls production and marketing through the ownership of capital and thus simply retains the surplus.  Wages are maintained below their actual value by the competition between workers for employment, and the monopoly power of the capitalists.  The working class is defined by the relationship of people to capital that requires those who are not capitalists to sell their labour, and the working class is necessarily exploited as a result of the capitalists need for profit.

The motivating force driving the capitalist is the accumulation of capital.  Each capitalists wants to accumulate ever greater concentrations of capital making it possible to establish production of an ever larger scale and at the same time ever increasing his or her wealth.  And with wealth come not only riches but power and privilege.

The underlying structures of society are thus the organization of the means of production and the division of society into classes.  This division is between owner and worker, exploiter and exploited, rich and poor.  Having the essential character of a commodity, whose services may be bought and sold, labour is always in a state of excess supply, creating a pool of unemployed labour, known as the reserve army of workers.

Capitalism is both capable of great growth and expansion and of formidable stagnation and decay.  Everything is dependent upon the rate of profit of the capitalists.  For a long historical period the rate of profit has been rising due to industrialization, cheap energy, economies of scale and scope, population growth, imperialism, and globalization, among others.  In many parts of the world this has even permitted wages to rise as capitalists have needed growing amounts of labour who accept the terms of the capitalist bargain.

However, the long term future of capitalism is not a good one.  As capital accumulates in ever greater amounts, the law of diminishing returns sets in.  The result is a falling rate of profit overall through time.  Capital accumulation cannot proceed unabated into the future without undermining the very basis for its own success.  Over the long term, capitalism is doomed to its own destruction.

Early Marxism had very little to say about the government or public policy.  Marx’s fundamental proposition was that the state is simply a committee of the capitalist class, dedicated to undertake measures to shore up capitalism as it works its way to its inevitable demise.  The government or state has no autonomy from the interests of the capitalist class; public policy is completely determined by the needs of capital.

According to this explanation, services such as education, health care, unemployment insurance and welfare are all necessary to sustain capital and the profitability of capital.  Had governments not introduced these things, capitalists would have had to make them available in order to sustain the profitability of capital.  Government permits the costs to be shifted off to taxpayers generally, thus not placing a burden on capital and a draw-down on profits.  The state, which constitutes another of the structures of society, shares the burden with workers of maintaining the rate of profit by exploiting individuals in society through taxation.

Marxism is described as structuralism because the structures of society determine outcomes within society as part of historical processes.  These are beyond the conscious actions or decisions of individual agents or organized groups.  The only group capable of consciously having an impact is the working class, who can accelerate the decline of capitalism by organizing a revolution, and displacing capitalist ownership of the means of production with worker ownership.  Such a thorough re-structuring displaces capitalism with communism.  However, revolutions are not the primary cause of this change.  Revolutions simply push the inevitable along at a slightly faster pace and provide the basis for putting the new structures in place, since capitalists will not readily give up their power or privilege.  And even revolutions must await the historical development of capitalism to a point of relatively full maturity, at which point the rate of profit will approach zero, the process of industrialization will have completely invaded economic life, those without capital will be largely incorporated into the working class, and capitalism will have exhausted its power and potential as a force for further increasing the material capacity of society.

Because the rate of profit is in long term decline, and the state without evening really knowing it must work to maintain the current structure, the state is constantly in a position of having to adopt new policies to arrest the long term erosion of the rate of profit.  The proximate cause of policy change is the falling rate of profit.  It is not necessary to have a model of how the state actually arrives at a particular policy, since all policies have the same cause and effect.  Other more complex explanations simply unnecessarily muddy the waters.

Critics of this approach have pointed out some difficulties, among other things, with squaring the facts with the predictions.  Capitalism has shown a remarkable resiliency, and if anything appears less vulnerable to decline and expiration than at any time in its history.  The rate of profit is not obviously declining over the long term.  Further the working class is declining as a proportion of the income earning population, and the ownership of capital has spread well beyond the capitalist class through public shareholdings.  And there appears to be something close to a consensus that capitalism is here to stay and must be either tolerated or actively supported if world poverty and the under provision of public services are to be addressed.

2.       Interest Theories

The political science and sociology literature contains a considerable amount of theory dealing with the role of organized interests in the determination of policy.  In some formulations there is a single group in society which plays the determining role in what policies are adopted.  These are generally people of superior education, status, wealth and power.  The members of this elite group share similar cultural, educational and socio-economic backgrounds, and tend to think about and analyze problems in the same way.

In some formulations, the power and authority of a single group takes on a structuralist tone, and have similarities to Marxist formulations.  The most notable example of this is the work of Lindbolm, a prominent United States political scientist.  After a long study of the role of groups in policy making, he concluded that business needs control the determination of policy, and that the over-riding imperatives are the profitability of  business and the maintenance of investment.  Governments need profitable businesses in order to provide growing tax revenues.  Anything that threatens tax revenues reduces the ability of government to provide programs and services, and increases its vulnerability with voters.   Business investment is needed to provide employment, maintain economic growth, and ensure economic stability.   This means that governments have no choice but to pursue policies that will maintain business confidence and keep investment flowing.  These imperatives have become essential features of modern democratic capitalism.  They have become an inherent part of the understanding of government officials and business about how public policy issues are to be framed.  They are simply assumed as givens, and thus will shape policy decisions whether or not they are explicitly formulated as part of a decision, and whether or not business interests assume a role in the policy process.  This is simply another form of structuralism.

A variation on the Lindblom proposition is that all democratic capitalist societies have within them certain people that make up the elite.  Members of the elites essentially determine policy.  Their preferences determine policy outcomes.  In the structural version of elite theory, the power of the members of the elite does not require that they be part of a particular organized group.  The transmission of policy preferences does not take place through group processes.  Elite values and beliefs are transmitted and embedded through cultural, communicative and educational processes that are common to the elites in society.  Given that members of the elites occupy the most senior positions in government, industry and educational and cultural institutions, it simply follows that public policy will reflect their values and beliefs.  It is not necessary for them to analyze beyond a certain point or to work together as an organized group, given a pre-disposition to understand problems and issues in a certain common way.  This encompasses an acceptance of logic and needs of capitalism and, the need to place the interests of capital first.

3.     Socio-Economic Determinants

Another once popular form of structuralism, quite different from Marxism or elitism, but nevertheless sharing the assumption that policy takes the shape and form it does, without reference to the actions of individuals or organized groups, is based on socio-economic determinants.  Policy, it is argued, is determined by quantitatively expressed social and economic variables.  These variables, which are proximate measures of the stage of advancement in economic terms, explain why policy changes as it does at a particular stage of development. This is confirmed by policy convergence among countries as the state of their development converges.

The best known and most representative of this approach is the work of Harold Wilensky (Wilensky, 1975).  Using the comparative policy approach, Wilensky studied social security policy in sixty countries.  The key observation of policy was the percentage of each country’s GDP devoted to social security measures, on the basis that social security policy measures would be essentially the same at each different percentage [check this and verify].  He examined changes in this variable, and thus in social security policies, relative to four independent variables:  the age of the social security system, age distribution of the population, the GDP per capita, and whether the state is democratic or totalitarian.  The strongest correlation, he found, is with the GDP per capita.  From his data, he concluded that social security policies converge, or become the same, at similar levels of GDP per capita.   He therefore concluded that most of social policy differences can be explained by the relative wealth of the country, and that policy changes are the result of objective conditions created by economic modernization and development, and not the result of the actions of political parties, interest groups, or political or bureaucratic officials.  According to his findings, it makes little difference whether groups or individuals are active in trying to address social security problems

Problems With Structuralist Theories

There are a number of problems with structuralist theories.  One is the form in which they are constructed.  It is often difficult to establish exactly how they would be disproved.  Take the case for socio-economic determinants where the cause is actually ‘proved’ through quantification establishing cause and effect.  But policy sameness is established without observing policy as it is normally understood.  The fact that the proportion of GDP spent on social security is roughly the same for counties with similar GDP per capita does not speak to actual policy.  Social security spending includes such very different things as pensions for seniors, unemployment benefits, welfare benefits and care of indigent peoples, and care for the elderly and the disabled.  Sameness and difference can only be meaningfully expressed by looking at these actual policies, not expenditures on them in aggregate.  But all of them are constructed through the use of language and arguments that stretch credibility, and applying a methodology that doesn’t conform with normal standards of verification.  The facts are made to fit the theory, other than the other way around.  In the case of the socio-economic determinants, the definition of a policy change is stretched and contorted to the point that it either explains nothing, or what it explains is so obscure as to be very meaningful.  In the case of Marxism and structural Elitism, while many of the arguments sound reasonable, it is hard to see why they are more reasonable than alternative explanations which fit the same outcomes.  For instance, policy may be constrained by the need to maintain economic stability.  But both Marxism and Elitism provide a very limited basis for establishing why.  Indeed it is not clear what facts would be used to prove or disprove the arguments one way or the other.  Thus there is a tendency to want to look elsewhere for better explanations.

A second and related problem is what the theories leave out.  None of them provide for a mechanism though which the causal forces actually work.  The state or government appears almost non-existent for policy purposes, becoming simply an instrument for organizing the delivery of policy and for the management of policy implementation.  And to the extent that government plays a role, there is no explanation of how the playing of that role works.  Indeed, there isn’t really any attempt to explain the role at all.  Government is simply a black box, pliant and unimportant.

Structuralism enjoyed its greatest acceptance in the 1960’s and the 1970’s.  By the 1980’s, many structuralists were conceding that the role ascribed to government was inconsistent with experience and observation and undertook to correct the latter deficiency by ‘bringing the state back in’.

Technology and Structure

There is now a general acceptance among Marxist scholars that the theory in is classic form is inadequate as a description of the relationship between the economy and the government or state.  One argument advanced is that, notwithstanding the relative permanency of structure as described by Marxism, there can be a degree of variability within the structure of the capitalist economy which call for policy responses from governments that are different than those previously.  The principal causes are technological shifts.  During the first mature stage of capitalism, a dominant feature of capitalist structure was the concentration of production in large scale industrial plants applying mass production techniques.   As a result of technology that had developed along with the rise of capitalism, economies of scale, large scale production, standardized products, concentrations of labour in centralized locations, mass advertising, rapidly expanding mass markets for low cost standardized  products, and concentrated ownership were critical features of the mode of production.  The new Marxist thinking acknowledged that product standards, labour laws to ensure stability in mass labour markets, and capital intensive infrastructure were required of a state system in order to stabilize the employment relationship and make mass markets work effectively.  The stabilization of demand through Keynesian policies, mass education and training, and the socialization of large scale risks all were needed.  If not undertaken by government, the profitability of capitalism would have been severely undermined, and its role in economic undermined .  However, with the state assuming these kinds of responsibility, capitalism was provided the opportunity to develop and prosper.

The assumption of these responsibilities by the state was often rationalized as meeting the needs of the people.  However, significant policy developments of these kinds were always assisting capitalists to maintain the rate of profit within the economy.  Without these government measures, that would have not been possible.

But all of this is part of an economy that is rapidly disappearing.  The adoption of the types of policies referred to in the above characterized a certain stage of capitalist development, which was coming to an end in the 1960’s.  At the time, technological changes were bringing about dramatic changes in production and markets, highlighted  by the slogans ‘small is better’ and ‘the information economy’.  Smaller scale production, more specialized products, decentralized production, and an increasing concentration on services, which are more localized in production and distribution, were assuming ever greater importance.

The old policy framework, with its dependence on extensive regulation and standardization, was becoming lees and less suited to the needs of capitalism.  New policies were needed to adapt to this changing situation if an emerging decline in the overall rate of profit was to be arrested.  The standardization and regulation of production and products, the regulation of labour markets, the provision of large scale infrastructure, and the regulation of the behavior of firms in markets all began to undermine maintaining the rate of profit.  Technology was changing, and the changes made it necessary to reshape the role of the state in maintaining profits.  Change is slow, and these changes are still underway in the form of de-regulation, reduced public spending as mass pubic services are reduced and modified, welfare and education reform, and the reduced investment in mass infra-structure.

An interesting aspect of this approach is the importance given to technological change.  Defenders of capitalism also see technology as the source of change and renewal under capitalism, but they see technology as endogenous to capitalism and as a never ending source of renewal.  Capitalism is driven by competition, technological innovation, entrepreneurship and the relentless pursuit of profit.  Competition for labour and increasing productivity of labour arising from knowledge and skills drive up wages, and the fruits of capitalism trickle out into society and make all people better off.

Perhaps the most important observation about this work by the new Marxist theorists is the new attention to government and policy change.  The state of government is no longer unimportant in the scheme of things.  It remains, however, a passive instrument for implementing what is needed by the capitalist class.  It is not a vehicle in itself for mediating class conflict, directing economic and/or social transformation, or for autonomous or independent action by state actors.

More generally, a trend began in general to question strucuturalist theories because of the absence or irrelevance of the state in their common formulations.  Governments do regulate capitalism, often over the objections of the leaders in the business community.  Government does pick and choose winners, often becoming embroiled in class and other kinds of political conflict.  And government does appear at least to select policies that transform social and economic conditions, contrary to the direction preferred by the capitalist class, and that differ for reasons that are not entirely clear.

An even more fundamental problem for Marxists has been the stability of capitalism.  Surpluses and accumulation do not appear to be in long term decline.  Class is becoming less and less clearly defined in Marxist terms.  Real living standards have improved for  others than just the members of the capitalist class in capitalist economies.  The spread of share ownership in capitalist enterprises has contributed to a spreading of the fruits of capitalism.  The pervasive dominance of markets is increasingly recognized as having greater structural explanatory power than the concentration of capital.  Capitalism is apparently not threatened either by a decline in the rate of profit or class conflict.

The above referenced problems have led to a decline in interest in structural theories.  The concession the state does have extensive autonomy and is thus not bound totally by structure, and the increasing dissonance between the predictions of Marxism and  historical reality, have tended to lead public policy scholars in other directions, in attempting to explain policy and policy change.

Institutional Theories

Institutional theories, contained within what is known as institutionalism, address the state and government in a quite different way.  Institutionalists argue that it is only by understanding institutions, and particularly political institutions, that we can understand what it is that fundamentally shapes policies.

Institutional theory has developed largely through the use of comparative studies and from these the determination of policy differences that are explained by institutional differences.  This has changed somewhat in very recent times as theorists have turned their attention to organizational behavior, and in particular how individuals behave in differing organizational settings.  Before examining this later work, however, it is important to first get an understanding of traditional insiottutionalism.

In so doing it is necessary to understand what is meant by institutions.  One might initially assume that the answer is simple and that institutions are simply the bodies of government, such as legislatures, cabinets, government operating agencies and departments, central agencies, independent regulatory bodies, and the like.  In actual fact this is not quite the way institutions are characterized.  Rather institutions are seen as embodied rules, prescriptions, and constraints that determine and control action.  It is this way of looking at institutions that makes room for a kind of generality that permits theories to be constructed.

These rules, prescriptions and constraints are categorized and differentiated as follows.  It should be noted that this differentiation is important because observations are based on cross-country or comparative studies, as has already been indicated.

1. The relation between the legislature and the executive, and in particular the distinction between parliamentary systems and a system of checks and balances and formal division of powers such as in the United States and many other countries with directly elected Presidents.
2. The relative power and strength of legislatures vs. First Ministers, Cabinets and bureaucracies.
3. Electoral rules and norms, determining how candidates are selected, how representatives are elected, how the executive is chosen, the roles that parties play, and the extent to which power is contested.  These have an impact on the stability of election outcomes and the extent to which outcomes tend to gravitate around the middle.  Important distinctions are made between alternating single party majorities, durable single party dominance, and multi-party coalitions.
4. Constitutional architecture, including constitutional conventions.  These result in differences such as federalism, bicameralism, and the extent of judicial review and the role of the courts.
5. Procedural aspects of legislatures, including the extent of party discipline and forms of legislative voting.
6. Political culture and political norms and values, including those that increase or decrease political differentiation and contestation.

Many of the findings of instittutional studies have fallen short of explaining differences in policy.  They have been substantially better at describing degrees of co-ordination within governments, and the extent to which varying degrees of co-ordination enhance or reduce the capacity to make policy changes.  They have been much less successful at explaining differences in the actual content of policy changes.

It should be emphasized that institutional studies as defined in policy studies do not involve simple descriptions, including differences, in the behavior of legislatures, Presidents and Prime Ministers, Cabinets, bureaucracies, and courts in terms of specific policy problems.  Rather they to define meaningful differences in the way these are established, controlled and function, in order to correlate these with policy differences as they apply to specific policy problems.  The observations of differences are expressed in the form of rules, conventions and prescriptions, which become the generalized characteristics of institutional differences.

A seminal study of institutional differences draws the following conclusions:

1. Institutions affect differences in government capacities to make and change policies, rather than determining differences in policies themselves.
2. Although institutions affect government capabilities, their affects are contingent.  In other word, institutional differences will only explain differences in capabilities if other things are equal.  In addition, a specific institutional arrangement is neither a necessary nor a sufficient condition for a difference in policy capability.  Thus in certain cases it will provide opportunities for policy change because it enhances the capability for change, but only if other conditions are also present. “for example, party government-style parliamentary systems are most likely to provide opportunities for policy innovation when the governing party has a strong majority and cohesive political leadership and faces weal electoral threats from a divided opposition”.  (Weaver and Rockman 1993 p. 447)  But it is only the capacity that is enhanced; the actuality is still another matter yet.  The connecting link is tenuous indeed.
3. Specific instiutional arrangements often create both opportunities and risks for individual governmental capabilities.  In other words, a particular variable may influence policy change in either direction.  A high degree of insulation of a large parliamentary majority from other variables may mean that the majority is indifferent to pressure from interests, or indifferent to opposition from opposing interests.  It all depends, but apparently not on whether the system generates large stable political majorities.  If this is a relevant variable at all, it is certainly not very efficient or robust.
4. Policy making capabilities may vary across policy areas in the same institutional system.  This finding is not particularly problematic in that it is simply a re-statement that the influence of a single factor depends upon other things being equal, although it would be interesting to know what the other key variables are and how they interrelate with institutions.  For instance, if it could be said that if a, b, c, and d are the same, then in all policy areas under a particular insti arrangement policy capacity will be increased, or policy variability will be increased, or even better policies of a certain general type will be adopted, it would be possible to understand how insti  arrangements make a difference.
5. Some of the specific “other things” that can affect the impact of institutional arrangements are elite cohesion, multiple veto points, elite stability, interest group access, and elite autonomy from short term interests.  These are selected out form other variables because these in particular define a kind of state independence or autonomy from institutional arrangements.  There is a suggestion that this autonomy is a result of the state being able to shape a will or direction of its own, because it has the capacity to make decisions regardless of what the institutional arrangements.  However, the implications of this suggestion are never fully explored in traditional institutionalism.
6. Parliamentary type and U.S. type division of powers do not have each a unique affect on policy capacity, even with other things being equal.  The parliamentary system tends to concentrate power, while the U.S system diffuses power and requires more attention to coalition building.  But in the end each has the same overall affect on capacity, creating the same kinds of opportunities and constraints.

An important point to acknowledge is that insti. studies today largely share the view that isnti, are important in the sense of creating opportunities and setting constraints on the capacity to make policy changes. Nothing in general can be said about which of these they do.  Many if not all institutional arrangements do both.

In the final analysis, institutionalism is that it tends to depend upon capacity to make change.  Particular arrangements may enhance policy change or constrain it, depending on the circumstances.  No suitable explanation is provided for the subtle but important focus on capacity for change rather than change itself.  In the final analysis, one can only conclude either that differences in policy do not depend in any meaningful sense on differences in institutions, or that the impact of differences in institutions is affected by so many other factors that it cannot be described and substantiated within the current state of theory.

One apparent difficulty is the independent existence of the government or the state, in the sense that the government always has the ability to choose.  The so-called autonomy of the state raises its head again.  In the absence of a model of state behavior, it is virtually impossible to describe how institutions impact on policy changes.  Recently there has developed a new institutionalism that in part at least tries to remedy this problem.  An examination of this development will be deferred until the question of decision making is re-examined in its own right.

Group Theories

A popular form of theorizing is to attribute policy change to the organized activity of groups within society.  Many groups in society are organized for the purpose of pursuing public policies in accordance with the interest of their members. It is these groups and their activities to which one must look to understand policy outcomes and policy changes.

The study of groups has a long and honourable history in political science.  The most influential of the early work was pluralism.

1.       Pluralism

Pluralism begins with the observation that policy is determined through the interaction of organized groups in society.  Society has embedded within it, as a result of the situation of individuals and of social and economic circumstances that have developed, a large number of interests, some of which are competing and some of which are complimentary. Interest groups are organized around common or complimentary interests.

The fundamental goal of interest groups is to realize policy outcomes favourable to the interests of the members of the group.  For many of the early pluralists, society was nothing more that organized groups.  Groups can be formed and dis-banded as the circumstances demand, and people and businesses can belong to more than one group.  The groups internally provide for a certain degree of conflict resolution, since it is unlikely that those with common interests will see and understand their needs as being completely identical.  Common and over-lapping membership in different groups also helps to reduce conflict between groups.  But the real work of forming policy happens when the groups interact with one another.  The problems inherent in co-operating and reconciling winners and losers are resolved through groups and their interaction with one another.  The resolution of the essential problems of policy making falls to groups and group interaction.

The role of government in policy making becomes essentially neutral.

Government has no interests of its own, and so to the extent that it has a role it is simply one of providing a forum through which groups can interact.  It may to some extent mediate the competition between groups, but in doing so it is nothing more than act as a neutral party.  In the traditional and original formulation, access to the policy process is open, groups are assumed to have power approximately in proportion to the importance or intensity of their interests, and all interests are represented by groups.  The countervailing power of groups ensures that no one group dominates.  On the basis of these assumptions and observations, it was concluded that the interaction between groups provided a relatively accurate aggregation of preferences of individuals in society.  The plausibility of pluralism was enhanced by the fact that it seemed to confirm the experience of the United States, where organized groups have a powerful and observable role, and where predictions about policy outcomes seem to be possible by observing the engagement by organized interest groups.  Indeed, studies of the behavior of groups in the formation of policy gave rise to a focus on behavioral policy studies, and that had not existed until then.  It finally seemed possible to subject policy studies to empirical methods that had not been possible previously.  (**CASE STUDY:  SOFTWOOD LUMBER)**

Pluralism came to be quickly challenged as an over-simplification of reality.  The early challenges to traditional pluralism were threefold:

1. A questioning of the assumption that all interests are represented in the policy process by groups.
2. A questioning of the assumption that power and influence is distributed in proportion to the intensity of interests.
3. A questioning of the assumption that the ground upon which groups operate is neutral ground, and that government is essentially absent form any active role.

The first of these challenges led to an adaptation of traditional pluralism through the inclusion of potential groups.  These exist in the from of knowledge of the existing groups that there are those with shared interests who are not actually organized.  But they can and will organize if existing groups sufficiently threaten their interests (Smith, 1990, p. 305).  This possibility has the result that the other groups act as if the potential group is present and accounted for.

The second of these challenges lead to an acknowledgement that power and influence is not equitably distributed among groups.  Pluralism was gradually adapted to take into account differences in financial resources, in social status and credibility, in leadership skills and abilities, and in size and even access to the process.  And while it was acknowledged that his could in some sense bias the policy outcomes relative to the preferences of individuals, it was argued that this does not undermine the basic proposition that policy is then result of the competition and interaction between groups, including potential groups.

The third of these challenges came from scholars who were not satisfied with the treatment of government or the state in the pluralist description.  Once again, there came to be pressure to ‘bring the state back in’.  The assumption that government was not a player, or was neutral, was not accepted.  The first reaction was to treat government more or less as a mediating body, facilitating the resolution of conflict and ensuring that all interests are given a role appropriate to their interests.  The latter was challenged immediately on the basis that government is part of the formal structures of conflict resolution, and such formal structures are themselves shaped by interests and pressures.  These formal structures include in part the rules of the game, and these rules can mean that those who do not accept them are excluded or have their legitimacy reduced.  Government represents not neutrality, but the same kind of inequalities that exist in society.

In response, an attempt was made first to square the circle.  It was argued (Truman, 1951) that government is both an active and a neutralizing force, the latter because of pressures to ensure that the game is played fairly.  The bureaucracy tends to be organized in parallel to interests in society, and one part of it will not let another pursue the exclusive interests of its client group.  In addition governments require voter approval for reelection.  Voters will punish governments that do not ensure that groups are treated fairly.  In the extreme, voters could become dissatisfied with the system itself on the basis that counter-claims are being trivialized, and governments are not interested in a disintegration of the overall consensus supporting the system.

It has been argued with force that this still does not solve the problems with traditional pluralism.  Smith (Smith, 1990) has argued that it:

1. does not address the organization of policy making, giving rise to the need for an institutional analysis of policy-making.

1. does not address the role of ideology, which is equated to ideas, including ideas about the appropriateness of government intervention, its extent, and the means.

1. Does not take sufficient account of the external constraints on government, and particularly global factors such as international markets.

1. Assumes that access is more open then it really is, and that it has a greater impact than it often does.  Some groups are denied access for a very long time in some policy sectors, and in other sectors consultations have little impact on outcomes.

1. Assumes more influence of potential interest groups than really exists, and assumes more countervailing engagement by the bureaucracy than  actually exists.

The most telling criticism of pluralism and its claims of dispersal of power in society and the construction of a relatively balanced consensus comes from the structuralist perspective of interest domination already referenced.  Lindbolm, a long time pluralist, came to the conclusion that one interest is dominant in policy making, to the virtual exclusion of all others.  Pluralists have come to refer to this as neopluralism, although it is more akin to a wake than a revival.

The neopluralists recognize business groups representing corporate interests as having the dominant position in the contest between interests.  This is because government needs a stable and successful economy to achieve voter approval, which in turn means that business occupies a position in government that is qualitatively different (and infinitely more powerful) than any other group.  In addition, much of the corporate sector and what it does is beyond the reach of government, in accordance with values, conventions and understandings about how the world should work, and what would happen to profits and investment if government should intervene.  As a result, governments cannot consider validating policy requiring an incursion into that which is off limits according to these understandings.  Government actively constrains policy choices to protect special privileges of business.

The result can hardly be accurately described as a modification to pluralism.  Rather, is is a conversion to a form of structuralism, as has already been described.  The dominance of a particular interest is an expression of the structure power in society.  Its dominance does not depend upon a conscious and explicit decision to bow to corporate interests.  As Lindbolm says, “(A) government official simply understands, as is plain to see, that public affairs in market oriented systems are in the hands of two group leaders, government and business, who must collaborate and that to make the system work government leadership must often defer to business leadership” (Lindbolm 1977 p. 178)

In the face of business interests, government is neutered.

A number of variations and modernizations of pluralism have been attempted by a number of scholars.  Some of these involve attempts to provide more robust descriptions of the differences in kinds of engagements that occur between organized interests, and differences in the structuring of these engagements.  These will be addressed on the next section on policy communities and policy networks.  Others have involved an attempt to ‘bring the state back in’, in order to recognize that government is an active force in policy making.  They largely involve a recognition that government is part of a decision matrix in which it is engaged with groups to build an acceptance of a policy option. The degree to which it structures the rules of the game, and that there is a reconciliation of conflicting interests, and thus that various interests are taken into account, is in part dependent upon the preferences of government as it operates in policy sectors.   However, pluralism has very little to say about how government preferences are developed and how they operate.  Once again, the attempt to ‘bring the state back in’ comes up largely empty.

2.       Policy Community and Policy Network Theories

The policy making process has often been recognized as a multi-party process that is much more complex than the simple competition between interest groups suggested by the pluralists.  In the pluralist conception, interest groups bargain with one another with government acting on occasions a least as a neutral forum for the negotiations.

The concepts of policy community and policy network introduce a much richer formulation of who is involved and how in policy making.  Each explicitly recognizes that government is part of the interaction that occurs, that government brings certain preferences to the process, and that the number of government agencies involved may be numerous.  The groups involved may be many and varied, informed by very different ideas, or limited in number, similar in make-up and express similar ideas.  The ‘forum’ that defines who participates and the amount of variability is referred to as a ‘policy sub-system’.  This has similarities to a policy sector, but any given sector may have within it have differing policy sub-systems, depending on the particular isse.

A policy community is an institutionalized form of pressure group-government relationship which favours certain interests.  The insti. structure is formal, taking the form of an advisory committee, a commission, an authority, or what the British would call a quango, which is an arms length management, policy, or regulatory body that enjoys some independence from elected officials and their appointed agents, the regular civil service.  They are generally appointed to oversee the management of a particular issue or problem.  The structure will validate and perpetuate a certain dominant set of beliefs about issues, and there is thus a shared understanding about the appropriate analysis of a problem.  These beliefs and analysis for a particular policy community are defined and contained largely through the exclusion of certain groups and certain issues.  Under these circumstance all the members agree on what the policy options are, and what options are excluded.  The main activity of the community is to reach a bargain on the acceptable option from the limited options available, and to oversee the implementation of that option.  Bargaining is very limited, with trade-offs being the small trade-offs needed to reach an agreement amongst those with very similar interests.  A policy community is made of members with little diversity of interests, a shared analysis of the relevant issues, and who view the options for policy change in very small incremental terms.

A question arises as to the source of the rules and constraints that determine which dominant ideas will be the prevailing ones, and who can participate and who cannot.  After all, in many cases the range of potential ideas which could be relevant and of participants who have an interest is much larger than those actually included.  It has been suggested that this is structural, happening without conscious decisions (Smith, 1990, p319).  But this cannot be the case, even though it might appear to be by the participants.  The design of consultation processes clearly can be altered, and design changes of this sort are not beyond the will of at least the actors who participate.

The answer is that in order to find the source of the design, we must look to the state or government.  We must bring government back in, and assign to government the responsibility for design.  Smith agrees but goes on to argue that the problem is not entirely solved because of the diversity of interest of state players.  He insists that the state cannot be brought back in unless we mediate its return through the concept of interests.  As he says, “Different individuals, and groups of individuals, from various parts of the state have their own interests which they will try to promote”.  (Smith, 1990, p. 320).  Electoral officials have interests that are electoral, career, and ideological.  Unelected officials want promotions and increased power, they want to continue and expand their departments, and they have policy preferences of their own.  And departments and agencies have distinct collective interests, differing in at least some respects from those of officials.  In these circumstances, while groups are playing by rules set down by government, these rules themselves are a complex amalgam of internal government interests.  Further, given the nature of government, even the choice of the constraints on participation or on acceptable ideas is subject to pressure from interest groups.  If interest groups engage to try to get the policy they want, they will surely engage to try to get a structure that will facilitate getting the policy they want.  It is certain that interest groups will pursue government to get their preferred policy community structure.

One possible distinction in structure is that between relatively open and closed policy communities.  Open communities exist when a broad range of interests and ideas are included, in accordance with the rules which establish their legitimacy in participating.  Closed communities exist when the number of participants is very limited, and when a substantial number of groups with a direct interest in the outcome are excluded.

The case of open communities, it can be postulated that they will be observed when interests are diverse, and none enjoy privileged access to government.  This is because  power itself is diffuse and widely dispersed.  In the case of closed communities power is concentrated in one or a small number of groups.  This concentration of power arises from the amount of resources a group has, the skill and education of its leadership, and its skill at mobilizing public opinion.  It also depends upon whether a policy will alter the actions of the members of a group in such a way as to cause instability, and particularly economic instability.  Powerful groups can largely dictate how open a policy community shall be.  They do so by exercising power over government.  But it is government that in the final analysis makes the determination.

One can question whether or not it is true that in cases where there is a concentration of power in one or a small number of interest groups policy communities are always closed, It could be that in an exceptional case a powerful interest may consent to open up the community to others less powerful.  However, there is at least anecdotal evidence to suggest that exceptions are more broadly defined.  This evidence suggests that governments have more freedom to open up the community to interests and ideas than is suggested by the theory of the policy community (which is essentially an expanded form of pluralism).  Explaining this means that we must account for a more robust state than the theories  suggest. This we will have to be set aside for later consideration.

Many policy scholars make little if any real distinction between policy networks and policy communities.  Others make forced and arbitrary distinctions, with the result that there are a dizzying array of implied or stated definitions in the literature.  For our purposes, a policy network will be defined differently from a policy community in a small but important respect, namely that it is not constituted and formalized by government, and as such government does not determine the rules according to which a network functions, including who may join.  Networks are made up essentially of those who chose to get involved and who seek out ways to have an impact on government policy.  Members may or may not have a relationship with other non-government members of the network.  Any person or organization who attempts to make a difference to a policy change is part of the network.

An important resulting difference between a policy community and a policy network is that a community will often become involved in bargaining with government.  Because its members have power, they are able to provide something that government wants.  The negotiations in an open policy community will be multi-partied, with the government itself being a key party with favours to distribute and costs to impose.  Such negotiations can become very complex, and support for outcomes is not always particularly stable.

Networks don’t provide a framework for negotiations with government unless they are converted to such a form through government recognition and provision of rules.  When this happens, its members become part of a policy community, with membership now having to be sanctioned.

Networks are usually very open and quite changeable.  They end to have members with strongly opposing interests, the most extreme of which have little room for trade-offs.  Networks tend to be characterized by coalitions of like-minded members who organize to make their side of the case heard.  Coalitions form networks who do so are often referred to as ‘advocacy coalitions’ (Sababatier and Smith 1993).  Advocacy coalitions can include individuals and organizations from inside and outside government who share policy goals, causal understandings and beliefs about how government and society inter-relate.  As would be expected with members of groups generally, these coalitions have common interests and a shared knowledge or understanding of the policy problem.  There is a special recognition of knowledge, because they emphasize solutions.  There is also a suggestion of focused and coordinated activity and an energetic pursuit of ends.  In many respects advocacy coalitions are interest groups who have got their act together.  The distinction while sounding flip is by no means trivial.  The difference between success and failure in getting government to adopt a favourable policy is often dependent on successful coalition building and mobilization of resources and skilled people sufficient for a campaign.  Government may and often does respond to good a campaign that appeals to the media and the public, and that has drawn on people internal to government to advance the cause.

Groups and Power

Groups are important to policy outcomes.  Groups essentially exist to aggregate and express the preferences of their members.  Preferences are usually so similar as to be readily accommodated, and thus groups are able to provide a statement of preferences distinguishable from those drawn from a more broadly based population.  In order to ensure that its preferences are taken into account, a group also needs to distinguish itself in terms of power.  Power becomes an important concept when groups are considered as factors in policy determination.  Regular reference is made to influence or pressure by interest groups.  Influence exists when the behavior of others may be changed by the one with influence.   Pressure is when influence is mobilized.  Neither can exist without power.  To have power means to be able to confer benefits or impose costs of one kind or another on another party in such a way as to alter the behavior of the other. In the case of government, benefits may take the form of public expressions of opinion and support, greater cooperation on otherwise unrelated matters, assistance in elections, or simply praise that is credible in the eyes of the public.  Costs include reduced public support, actions which interfere with other things government wants, organization of opposition, public criticism, support for opponents in campaigns or any number of other things that detract from the government’s ability to achieve its goals.

But the groups are only one part of the relationship.  The other is government.  What government does and how government functions in the course of policy determination is still in need of explanation.

Systems Theories

On of the difficulties in coming to understand how it is that policy is determined is the complexity of the environment within which policy is made.  This problem has already been referred to, in the context of developing general theories of cause and effect.  A similar problem is the way that causes and effects themselves interrelate, and the way that social behavior becomes differentiated into different bundles of behavior, all or which impact on one another.  We can think of the overall universe of policy determination as being a number of behavioral locations, each responding to causes in certain ways but each also having a kind of free will of its own.  In addition each location and the behaviors within it is being impacted by each of the others, at times being a cause of something another does and at other times being an effect.  There are also external forces that impact behaviors

A Canadian political scientist, David Easton, (Easton, 1965) undertook to provide some order and clarification of causes and effects by defining what he terms the political system in a country or nation.  Five concepts are used to describe the system:

1. Inputs

The inputs in the political system are the ordinary citizens and the organized groups within society.  These themselves are of two types; those that provide support and those that place demands on government.  Demands and support can get channeled directly but most often indirectly through interest groups and political parties.  Support is provided through public expressions, help in elections and various forms of cooperation with government.  Demands can be expressed in a whole variety of ways, including through direct communication, public expression, pressure through the exercise of power, election activity and interference with government in proceeding with its intentions if contrary to the demands of a group.  Interest groups are organized around relatively common interests, whereas political parties involve larger coalitions where conflicting interests are arbitrated so as to make it possible to contest the objectives and intentions of others whose interests diverge to a more substantial degree.  The activity of groups and individuals is defined in part by laws and in part by political culture.  Political culture includes a whole range of norms, standards and conventions which are not written down but which are widely accepted.  Action and intentions outside of these are popularly considered as illegitimate and the outcomes invalid.  This ethic is given force through social sanctions.  Within the political culture are such things as the relationship of the individual to the state, the priority given to the state and state power, the role and recognition of popular expression, attitudes regarding authority and autonomy, and the degree of tolerance of dissent and difference.

1. Decision Making

Decision making is a process.  The process transforms inputs into outputs.  Decision making is located within the state. Since the state is all of the individuals and institutions that exercise power, the state in system theory appears to include all individuals, groups or organizations able to impose loses or benefits on others. While it does include government, and many relationships that according to group theory are mediated through government, in most formulations of system theory the initial characterization of the state appears to go beyond government agents and agencies in the public sector as we have defined it.   The possibility that a systems definition of the state might include cooperation and conflict in the non public sectors is interesting, but most systems theorists give this little real attention, since inputs are identified placing demands on and providing support to governments.

However, it is not uncommon in social theory to define the state more broadly than government agents and agencies.  This is done because of the desire to include within the state not just organizations but government relationships with private individuals and organizations.  Thus it is not just entities but relationships of a certain type that make up the state. This inclusion is a useful clarification, but it does not make the private individuals or organizations part of the state. The state is made up of those agents and agencies possessing the power to coerce.  The desire to present the state as a system is useful in clarifying the importance of relationships and this flows between the private and public sectors.

The most important observation about systems theory, however is that of decisions of the state as the source of policy.  It is these decision locations and what flows from them that are of greatest interest, since it is through decisions that inputs get converted into outputs.  They are the processing centers, and processing of inputs into outputs is fundamental to any systems theory.

There may be strong states or weak states.  Strong states have the capacity to consider a wide range of policy choices, to respond to changing conditions, to make decisions quickly and efficiently, and to provide relatively stable and consistent policy outcomes.  They also have the capacity to implement decisions successfully, notwithstanding that there may have been strong contestation at the time the decisions were made.  Weak states have the contrary characteristics, not surprisingly since weak and strong are relative concepts.

1. Outputs

The outputs of the process are public policies.  We have already discussed the character and nature of public policy.

1. Feedback

The concept of feedback is perhaps one of the most important contributions of systems theory. It means that knowledge and information about a policy change itself causes reactions in people and organizations that puts in play further and additional inputs.  The idea is that policy change is itself in part a cause of further policy change.  A policy change causes people to react and to think and to refocus how they express their preferences and how their preference expressions are translated.  Even if preferences are stable, people and organizations learn.  Governments also learn as they gain more and more and experience, and this too causes changes in policy preferences within government.  Feedback causes change, and this change fundamentally reflects learning.  It is for this reason that change in policy is not often made in great leaps.  Very often policy changes involve adjustments as feedback has its impact.  But just as policy change reflects feedback, it does not imply instability or inconsistency in preferences.  Decisions require information and knowledge, and new information and knowledge flowing from past decisions will have an impact on current decisions.

1. The Environment

The environment is those factors which are determined by forces outside the system.  It can be thought of as part of a picture or atlas, in which those factors originating outside the boundaries of the system are part of the environment.  A system is in a sense bounded in which everything inside the system is related to everything else.  A change in one thing will cause a change in everything else.  Systems have the potential to reach equilibriums, reached when nothing inside the system changes, and all reactions to earlier changes have worked their way through, although feedback loops can mean that no final equilibrium is ever practically possible.

But there are nevertheless always factors which helped define the system originally, and which if changed, will “shock” the system into another round of changes.  These are exogenous to the system.  Changes in these exogenous factors are not caused by changes within the system.  These exogenous factors are part of the environment.  One is preferences of individuals, but these are assumed to be relatively stable.  Another is history, which also is unchanging.  Each of these play a major part in defining the system, but neither is though of as open to change.  (It is interesting to ask whether changes in the ‘salience’ of choices are exogenous or endogenous).

Other exogenous factors are economic, social and political structure.  These do not change as a result of decisions made within the system.  Natural endowments of a country are also exogenous.  So too are global conditions, which are all factors arising outside the boundaries of a country.

Functionalism

It has already been noted that system theory emphasizes stability and equilibrium in the absence of shocks from the environment.  System theorists turn to functionalism to explain this orderliness and stability.  And it finds it within the system as a whole and not in the actions of individual agents within the system.  Functionalism is different from most social theories because these are rooted in the individual and the behavior of individuals as the starting point.  From a functionalist perspective key relationships are derived from the behavior of individuals and the aggregation of actions of individuals.

Functionalism approaches an understanding of relationships differently.  Rather then emphasizing changes in behavior and changes in outcomes, it focusing on the overall stability and order of the system.  Its emphasis is the system, and thus it is integrally related to systems theory.  The primary observation of functionalists is that the political system resists disturbances originating in changes in the environment, and tends to maintain itself intact even in the face of relatively frequent events that upset the balance in the systems.  As has already been noted, in systems theory there is a tendency for the system to find an equilibrium and to move back to that equilibrium in response to disturbances from the external environment.

An obvious question is why this is so.  What is the explanation of stability and order.  Why does change result in small adjustments, and why do the feedback loops contribute to stability rather than to ever and ever greater change.  The answer is that institutions in society have functions, and those functions can be defined in terms of their contribution to adjustments to a stable equilibrium.  Functions are defined in terms of the contribution to the maintenance of stability.  Indeed it is these that explain the development of institutions and the form they take in society.  Systems have the character of self- preservation.  Each institution or entity has a function, and that is to maintain the system and the stability of the system.

For functionalists, the important empirical question about behavior is how it contributes to the maintenance of stability.  So too with institutions or organizations.  The idea of the autonomous state or individual is based on a mistaken interpretation of society.  The organizing principle of social organization is that institutions develop and exist to maintain order and stability, and each has a contribution to make in this regard.  The individual is simply an agent of the system, not the other way around.

This description naturally raises the question of design.  If this is an accurate description of society and thus of the policy process, how did it come to take on this form rather than some other?  In reply, functionalists tend to turn to biology, and point to such things as animals.  Animals are constituted as a system with organs, circulation flows, nerve transmitters, and feedback loops all contributing to the maintenance of the body in equilibrium in terms of heartbeat, temperature, cell survival and a whole host of other things.  Each organ has a function, and the function is to keep the organism in a stable equilibrium measured in any number of ways.  And importantly, not all functions are carried out by a single organ.  There can be overlaps in functions.  This kind of overlap is called redundancy.  Indeed redundancy is a key feature of functionalism, in that all systems have certain organs that will step in and take up the functions of another if it is to fail for any reason.

In the case of animals, we do not have clear answers about design.  Some attribute design to God, although this not accepted by scientists as adequate.  Others attribute design to evolution, in which certain mutations are found to be better fitted to the environment, leading to an adjustment to a new equilibrium, but which then becomes stable and persists.  Others belief that evolution cannot answer all of our questions about design, but emphasize that this still does not prevent from observing that the key feature of organisms is the functional nature of the organs and relationships between them.  And these functions are all derived from the imperative for order, stability and survival.

A problem with functionalism is that it draws an unjustified parallel between organisms and societies.  It suffers from the problem of reasoning by analogy.  Humans and social organizations have the capacity to make choices.  Free will is a defining feature of human existence.  Cells and organs, no matter how complex, do not have this capacity.  They do not reason and they do not make decisions as a consequence of such reasoning.  The analogy between organisms and society is interesting, but it breaks down as soon as we look at interesting systems such as policy systems.  Easton defined decision making as a key element of the political system.  This having been done, it undermines not only functionalism but some of the key assumption of systems theory itself.  If relationships and feedback loops maintain the system, the question remains as to why this is so.  It cannot simply be attributed to the observation that there is a tendency to stability, and that therefore that must be what they do.  If the observation is valid, there is still a question where human agencies is involved as to how and why the agents choose to behave this way.

Systems theory is no longer generally thought of as a useful theory to help us in sorting out how any particular policies get made, nor does it add much in terms of generalized descriptions of policy making.  In many ways it is simply another form of comparative studies.  The concept of feedback is important, and so too is the importance placed on decision making.  However, the characterization of policy decision making as a production process has probably done as much harm as good.  Not only has it supported functionalism, but it has contributed to the idea of policy making as a conveyor belt of inputs proceeding along to be combined and processed, with feedback loops in the processor dumping certain downstream outputs back into the process as further inputs.  This latter is referred to as the stages model of policy decision making.

The problem is that such characterizations quickly turn to models that attempt to confine decision making to regularized stable process.  They appeal to a sense of regular flow and order at the expense of observing what is really happening.  Systems theory has added only a limited amount to our understanding of decision making itself.