Public Policy

The public domain is best explained by definition that economists give, especially in terms of the role of governments in the free market system.  In Economics & Public Finance we've learned that the role of government is to correct market failures.  Market failures occur when there are externalities, that is, consequences of actions from the market and inequality.  But more importantly, economists see the government's role in the market to provide public goods.

Although there is much debate on what public goods precisely are, it is generally goods that are available to all, produced by the state and non-excludable.  In the private sector, there is a direct relationship with operational priorities and maximising profit.  The public sector on the other hand, is not governed by markets, but rather by political decisions on "economic & societal priorities" (Parsons, W. 195: 10).  This is briefly the domain of public sector.


The objective of this article is:

  • To understand the concepts and framework of policies.
  • To understand the public policy processes.
  • To apply different models and basic steps in analysing policies
  • To understand the monitoring and evaluation of policies
  • To understand public policies in a developmental context.

 Public Policies

Many different definitions exist for public policies.  A very concise definition of public policy will be that of a course of action or intention by government to do or even not to do something.  The CDE define public policy as a purposive action based on currently acceptable societal value, in dealing with a problem, focussing on the role of the current values of society.  Taken form this, I would define public policy as government's intentions and actions, through its agents, to address a perceived problem or to impose government's economic and societal priorities based on their current values.  This definition recognises the fact that government is an important actor in policies and policies are therefore the actions or intentions of government.  It also recognises that the implementation is through the governmental institutions.  Most importantly, it recognises the fact that current values and priorities are important in understanding public policies but also points to the possibility that public policies can address perceived problems, even issues which promotes more the career of individuals than addressing a priority need (ex: Gautrain).

Proper public policy analysis involves much more than merely considering the dissatisfaction of the public.  Parsons (1995) referenced Wildavsky in Public Policy on the multi facets of policy analysis, and stated: "Policy analysis is an applied subfield whose contents cannot be determined by disciplinary boundaries but by whatever appears appropriate to the circumstances of the time and nature of the problem" (Wildavsky, 1979: 15).   The nature of policy analysis requires a multi-disciplinary approach.  I found the 6 steps  of Policy Analysis of Patton & Sawicki (1993) very helpful in understanding the study of public policies.

But even more than analysis, the models of policies opened my understanding of the many intricacies of public policies.  Dye (1995) explains the meaning of models as a "simplified presentation" (Dye, T.R. 1995: 18), or an abstract conceptualisation of the intricacies of a public policy.  As a tool for analysts, it provides a simple "terrain map", helping analysts to identify important aspects of policies and its problems, to be able to understand public policies better and to explain the outcomes or consequences of public policies. 

Over time, analysts developed different models of understanding public policies.  Because of the complexities of human behaviour and social science, not 1 model can fully describe the development and implementation of a policy.  Therefore, these models equally direct us to different facets of public policies.  Dye (1995: 18) identifies 9 models as conceptual frameworks of public policies with the aim to simplify our analysis of public policies by directing us to factors affecting policies.  These models are: institutional model, process model, group model, elite model, rational model, incremental model, game-theory model, public choice model and systems model.  De Coning (1995: 132) categorized policy models in 2 categories, descriptive and prescriptive models.  According to de Coning, descriptive models analyses the process of policy making while prescriptive models analyses the content, impacts and consequences of policies.

 Institutional Model

The institutional model is classified as a descriptive model and sees public policies as a product of public institutions.  Because the government have the authority to legitimise its actions, they can "command" loyalty to their policies and even punish policy transgressions.  Also, only government policies have a universal character, applying to all its citizens.  This model approaches the implementation of policies as a top-down approach. 

Its usefulness can best be described by the fact that a change in government not necessarily means a change in policies, because policies are a product of the institution.  This model provides a way of analysing the behaviour of different public institutions.  For ex; if an important policy was formulated by government and was supposed to be implemented by local government, the failure of this implementation can be studied by looking at local government bodies as the institutions of government implementing its policies.

The critique to this model is that it disregards the role different political actors play in the policy-making process.  Booysen wrote an article examining the important role the 3 power clusters play in South African public policies, showing that policies are not merely the product of institutions.

 Process Model

The process model is also categorised by de Coning as a descriptive model and sees public policies as a series of political activities or processes.  These processes are typically the identification of problems, agenda setting, formulation of options, legitimising of policy, implementation and evaluation.  This policy acknowledges the policy-making process as a process taking into account many variables typical of social science. model

Its usefulness is in the way it shows how decisions are made.  Not the content of the policy is the focus of study, but the process followed in developing and implementing a policy.  This model links closer to democratic and especially, participatory ideals because of the ongoing consultation as part of the policy process.

Dye's critique on this model is that more participation, more "process" could change a policy's content, but doesn't necessarily follows that.  To me, this critique is to relevate the importance of consultation and public participation. in the policy-making process. 

Most policies nowadays do follow this process.

 Group Model

The group model, another descriptive model, according to de Coning, assumes that the interaction of groups ís politics.  Policy is seen as the initiative of interest groups.  Individuals are only important actors in the policy-making process as they participate in groups.  In this sense, groups become the bridge between society and government.  The consequence of the group model is that groups can engage in a struggle with each other, competing for more influence and/or more members to tip the scale in their favour.  As a group's members and consequently its influence increases, so will the policy-decision move to their direction and vice versa.  In this sense, a policy is an equilibrium reached between the pressure of a group and the government.

Practical examples of this model are the role TAC play in putting pressure on the government to change its policy on anti-retro viral medicine.  Also, the unions as a pressure group in labour policy, and members of the alliance (SACP / Cosatu) pressuring government in policy decisions.

The danger of this model is that government and policies can constantly become a "respond to pressure" and politicians might eventually self join groups or form coalitions to counter the pressure of groups.  A further critique is that group interests can become overwhelming, negating individual interests. 

The usefulness of this model lies in its analysis of the impact of political activity, and especially pressure groups.  Also, to reach an equilibrium in society is often a sought after ideal and often act as checks and balances because no single group can constitutes as a majority.

 Incremental Model

The incremental model, also categorised by de Coning as a prescriptive model, sees public policies as the continuation of current or previous policies.  Reasons why this model is favoured are because it is seen as the less radical approach resulting in the least conflict.  Also, where there is a constraint in time, cost or available information, it is easier to build on an existing policy, adding a few modifications.  It is therefore seen as the conservative approach, using existing policies as the base.

The usefulness of this model is the possibility of perfecting current "good" policies.  It is also easier to reach agreement between conflicting parties.

The critique to this model is that many problems can't be resolved by merely incrementally add on to existing policies.  Although it is the less radical approach, it is also the approach which doesn't require any change and acknowledges investments in current policy structures and processes.

In the South African context, many policies followed this model.  As Booysen mentioned in her article, the new government had a window of opportunity after the 1994 elections, to implement radical new policies reflecting the new all inclusive dispensation.  But in reality, many apartheid-era policies were merely incrementally modified, for example the foreign trade policies with the EU, education policy.  They were only changed to reflect all of society.

 Systems Model

The systems model is an interesting model and provides a valuable tool for systems analysis.  It is based on a simple input - system - output model.  Input is the support or demand from the environment.  System is the authoritative function of the government and output is the resultant policy.  This model then present policy as the outputs from the systems through debates and proposals to meet demands.

Its usefulness lies in the information of the policy-making process and political dynamics.  It also takes note of the environment (impact studies) and transforms public demands into policies. 

Critique against this model is that it does not address power relations existing in the policy-making process and implies a logic linear process (input - system - output) of policies.  It also doesn't explain how public demands get converted into policies.

 Rational Model

The rational model is one of the few descriptive models according to de Coning's (1995: 132) categories of the models and has its roots in modernism.  This model sees policies as a conclusion after all options have fully been researched, described, calculated and quantified.  The resultant policy will therefore represent the maximum gain with the least cost.  To reach such a conclusion, all social, economic and political costs must be calculated so that each policy can be weighted to reach the best decision.  Dye (195: 28) points out 2 important guidelines from this perspective namely:

  1. the gain must exceed all cost and
  2. only policies which will be able to give maximum benefit over cost, are considered.

The usefulness of this model is the scientific approach to policies, clarifying the consequences and impact decisions can have on society. 

But that already is part of the critique against this model.  It assumes that all values are accessible and quantifiable to be able to make such decisions as the maximum benefit to society.  It is generally seen that such quantification of variables are impossible due to cost and time constraints.  More-over, the quantifying of social behaviour and values are impossible, rendering this model to be impractical in such decisions.  For example, how do you quantify "A Better life for All", or human perception etc.  It therefore requires a huge "predictive capacity" (Dye 1995: 28) from decision-makers.  The result of this approach is that these "maximum-benefit" policies will eventually focus on economic, i.e. cost-benefit goals rather than goals of the society at large.  This might then lead to favouring existing policies because all the infra-structure, personnel, implementation and administration costs are already accounted for.  It might also be very difficult to obtain such in-depth information which this model assumes, because of the reluctance of political players to release such research information for different reasons, for example the sensitivity of the information, or the danger of such information proving the failure of a policy.

I could fully link a South African policy example which might have been developed with this model in mind, but I'm of the opinion that elements of this model may be present in many policies due to the advantage of collecting relevant information through research, albeit then only in factors that are quantifiable.

 Elite Model

The elite model sees policies as the decisions, preferences and values of a small elite group which shapes the opinion of the masses.  It is the few individuals who have power who decide on the values and policies of the government.  Movement from the masses to the elite is possible, but is seen as a slow and often compromising movement.  Thinking about the revolutions in the past (French & Marxist), it might be that the masses revolted exactly because of the assumptions of the elite-model politics.  In essence, the elites share consensus on behalf of the masses on basic human values, human rights and democratic principles.

The usefulness of this model is to give a picture of the power plays in politics.  Many practical examples can be seen world-wide, from the American politics to local politics.  John Higley & Jan Pakulski (undated) wrote an interesting article on elite and leadership changes in liberal democracies.  According to them, a shift towards elites and leaders and the concentration of power to the expense of the non-elites in liberal democracies are evident.  Examples of these shifts are taken from the presidency of Reagan and the current term of Pres George W Bush.  "In the course of 2001, especially after 9/11, it dawned on observers of American politics that an uncompromising elite had taken over." (Higley, J. & Pakulski, J. undated: 10).  Booysen (2001: 125) also mention in an article the importance of the President and the President's office in public policy in South Africa.  This can easily be identified as an elite group affecting the policy decisions, even sometimes making decisions unilaterally.

As practical example, the USA democracy comes to mind.  Personally it is the least democratic democracy because of its indirect representation, resulting in the least voting power to the masses.  Therefore, it seems as if voters realise the "low likelihood of a single vote being able to influence electoral outcomes" (Somin, I. 2004: 2).  American politics, and therefore policies, probably operates according to this model.   

Democracy demands an informed electorate. Voters who lack adequate knowledge about politics will find it difficult to control public policy.  Inadequate voter knowledge prevents government from reflecting the will of the people in any meaningful way. Such ignorance also raises doubts about democracy as a means of serving the interests of a majority. Voters who lack sufficient knowledge may be manipulated by elites.  They may also demand policies that contravene their own interests. (Somin, I. 2004: 1)

A South African example of this model operating in policy decisions might point us to a decision of the previous president on SA's policy regarding the recognition of Taiwan and China, which was apparently taken without much consensus.  In some circles (Booysen 2001), GEAR is also seen as a policy created by the economy-elites with little participation from grass-roots level.

Many critiques can be brought against this model.  This model operates with the assumption that masses are passive, "ill-informed" (Dye, 1995: 27) and don't know what's best for them, and the elite know and will act to the best interest of society.  It further assumes that all policies will solely be for the benefit of the masses and that the elite have the nobility of no self-interest in politics.  This model clearly does not reflect the demands of the masses.  It further assumes that there is no conflict between the elites, or that the conflict is on unimportant issues, and that there are consensus on the big moral values of individual rights, liberty and democracy.

Two of these models were of specific interest to me because of my interest in paradigm shifts, or social changes as we've learned in Democracy & Development; the Rational Model and the Elite Model.


References to the articles on policy implementation and policy evaluation and the problems associated with evaluation (Anderson, J.E. 1994: 244) are very informative. 

Another useful article on policy implementation is Brynard's (Cloete, F. & Wissink, H. (Ed) 2000: 165) article on policy implementation.  Especially his discussion on the 2 approaches to implementation; the top-down and bottom-up approaches.  To me this relates once again to the modernist and post-modernist paradigms of society.  The modernistic approach would be to direct and implement actions from top-down where a strong hierarchical structure and adherence to it, is the norm.  In a time where we've seen the gradual change of paradigms, more and more questions the hierarchical top-down approach, resulting in implementation of a bottom-up approach.  Although times have changed, and in future might even prove that more approaches can be added, both approaches still have useful insights in policy implementation.

Brynard's 5C Protocol (2000: 165) of policy implementation acknowledges all the different factors that are part of the implementation of a policy.  They are:

  • Content - policy implementation purport many interrelated actions.
  • Context - no policy operates in a vacuum and therefore implementation happens in a context.
  • Commitment - all levels of stakeholder's must be committed in the implementation.
  • Capacity - if capacity is not available, it should be created, and includes leadership, motivation and commitment.
  • Clients & coalitions - all stakeholders.

These 5 factors do not operate in isolation of each other and are also not a linear process.  It is a process where these factors operate parallel with each other.  This implementation protocol will help implementation actions to be directed to eventually achieve policy objectives and to obtain the desired results.


NOTE! This site uses cookies and similar technologies.

If you not change browser settings, you agree to it. Learn more

I understand