In a multi-party democracy like we have in South Africa, and also in many parts of especially
the developed world, parties and leaders are voted into office on the basis of their policies.
A party which on the balance of evidence, captures the hopes of aspirations of the electorate
and citizenry gets the nod to implement such policies over a pre-determined period of time.
And if such a party deviates from its manifesto or fails to implement the policy proposals
that it sold to the electorate it might be removed from office in favour of a better alternative
or be given another chance if no viable alternative presents itself.
The African National Congress, the governing party in South Africa, will hold a three day
2017 policy conference next week, to review its public policy positions for the future,
particularly as preparations for the next national elections in 2019. Over the last few
months, the party has circulated its policy discussion documents as part of a national
dialogue, that unfortunately has been dominated by other unsavoury developments on the
national scene. These mainly focussed on the Gupta family’s undue influence on
government decisions. But a public policy process is a delicate exercise that requires
effective meta-analysis, problem structuring, methods and models to ensure that it stays on
track and produces the desired outcomes.
But what do we mean by public policy?
Essentially, public policy is concerned with those areas of our life that are normally
described as the public good, are provided by the state and are “available to all”. These
include areas such as education, health, welfare, public transport, policing and economic
policy. However, as Parsons (1995) has observed, the distinction between what is public or
private has become a bit murky since the nineteenth and twentieth centuries. As a result of
the battle for terrain, we now have private schools and hospitals, toll roads as well as gated
communities. Generally, we can agree that policy refers to a statement of intent.
Practitioners who are concerned with this area of work, who deal with social problems,
policy content, gaps, outcomes and impact are referred to as policy analysts.
Policy analysts are guided in their work by models, theories and frameworks which, experts
agree, are aimed at ensuring coherence, consistency, congruence and cogency, amongst
others. Policy analysis is located within any of the paradigms of positivism, post-positivism,
critical theory and constructivism. I will not discuss these paradigms due to time and space
But generally, the body of policy analysis can be a bit pedantic; precisely
because it is a science. However, an effort has been made, hopefully with a measure of
success, to keep the discussion as simple as possible. This is in terms of both content and
There are numerous proposals about the stages that need to be followed in a policy process.
These include the Administrative Behaviour, Decision Process, Planning and Uncertainty,
and others. But in the main, the policy cycle begins with a problem.
It then advances to problem definition, identification of alternative solutions, evaluation of
options, selection of policy options, implementation and monitoring and evaluation.
How are policy processes managed?
Policy problems are not always straight-forward. Dunn (2004) warns that many times the
problems are inter-dependant, subjective, sometimes artificial and often dynamic. Howard
Raiffa, cited in Dunn (2004) refers to three critical errors in research and policy analysis.
The first type error is rejecting the null hypothesis when it is true. The second is accepting
the null hypothesis when it is false. The third type error arises when time and resources are
spent solving the wrong problem.
A policy process can be managed through the descriptive, normative, symbolic, procedural
and prescriptive models. De Jong (1995) informs us that descriptive models assist in the
policy formulation process whilst the prescriptive models assist in the evaluation of impact
and outcome. Symbolic models are a bit more complex and involve extensive use of
mathematical formulae. They are not commonly applied in the structuring of policy
problems. Prescriptive models include rational, incremental and mixed scanning
approaches. Descriptive models include functional process, elitism, group, institutional,
systems, public choice and generic process models. Perhaps we can do a brief overview of
some of these models:
The functional model recognises that any public statement only becomes policy when it has
been translated as such by a government institution. Only at that stage can it be
programitised, resourced, planned and budgeted for.
The group approach happens when particular interest groups e.g. labour or employer
organisations, instigates for a policy position, such as working hours. The NEDLAC
mechanism in South Africa is a manifestation of this approach. The advantage of interest
groups lies in their numbers and collective advocacy. T.R Dye (1995) has noted that groups
also have advantage due to their organisational strength, internal cohesion, leadership and
access to policy makers. They can bargain and negotiate.
The elite model allows a small elite to determine for the ill-informed masses or what an
apartheid leader termed ‘unsophisticated majority’ what is of common good. However, the
elite are usually drawn from the upper echelons of society. They therefore represent their
class interests than they do the wishes of the majority of ordinary people. Also, the elite
approach is hardly revolutionary; it is usually incremental.
The rationale theory seeks to achieve maximum social gain. It argues that the benefit of a
policy must always outweigh its cost. For an example, this policy would not support the
envisaged trillion rand investment on nuclear power. Opponents argue, the adoption of
such a policy will benefit a few corrupt domestic and international syndicates and place
undue and excessive financial burden on future generations. The model requires good
knowledge of society’s value preferences, available policy options, potential consequences
of each alternative, positive cost-benefit analysis and selection of most appropriate policy
option. The cost in this instance goes far beyond the narrow monetary definition.
The systems model is based on the conversion of policy inputs or demands into public
policy and the resultant outcomes or outputs of such a process. As Dye notes, it portrays
public policy as an output of a political system.
We can safely argue however, that none of the above policy models is perfect. Each one has
its own advantages and limitations. The appropriateness or otherwise of each will be
informed by prevailing circumstances, which include, amongst others, availability of
information, skills set, level of stakes and timeframes.
What policy approach does the ANC follow?
The ANC uses a combination of models in its policy processes. Notably, these tools are the
elite, group, institutional and to a lesser extent, the systems model. The elite play a critical
role during the policy formulation process in the ANC. They are fairly skilled, educated,
knowledgeable, articulate and confident. First, they set the agenda and draft policy
discussion documents. Then they move on to ensure general consensus on their proposals
amongst their regional and provincial counterparts before documents are officially
When conference breaks into commissions, the uninformed delegates often choose to stay
out of such commissions. Instead, they focus on song and dance; placating the names of
leaders they prefer for NEC inclusion. This way they also avoid an unfair contest of ideas
with the elite. The battle of ideas is then left between the elite and interest groups, such as
members of labour, communist bloc and Sanco, as representatives of their communities. In
this battle, the elite usually concede one or two positions to the groups. But by and large ,
their proposals prevail. In the plenary, there would be isolated ineffective resistance, but
again the views of the elite will prevail. Once adopted, it is expected of deployees to
translate ANC policy into government policy and proceed with implementation.
But a second process will begin in government, involving wide consultation with industry
and stakeholders and the engagement of NEDLAC. When consensus has been reached and
cabinet has endorsed the proposal, a parliamentary process often ensues. So by the time the
proposals are accepted as government policy, they could already be a watered down
version of ANC policy.
So, what is policy implementation?
Who implements public policy? How is it effected? Why are some policies successfully
implemented, and others not?
Brynard, cited in Cloete & Wissink (2000) observes that early scholars regarded
implementation as an administrative function which happens after the policy has been
adopted. He notes that this view has since been debunked. He further argues that a new
theory of policy implementation is still under construction. In this regard, V an Meter and
Van Horn (1975) maintain that implementation is about pursuing policy objectives.
Brynard argues that policy implementation requires a good understanding of the 5 Cs,
which are content, context, commitment, capacity and clients and coalitions. More
importantly, we can argue, effective policy implementation should start during the
In practice, policy implementation is largely a province of senior government officials. It
occurs when policies are being put into effect to address a societal problem. It is often
implemented through a combination of legislation, institutional arrangements,
programmes, projects and allocation of resources. In government, policy is the
responsibility of the political head. But it is senior managers in government departments
who are critical in the implementation of policy. Sometimes the private sector plays an
important role in the implementation of policy.
There are many reasons why policies are not implemented. Critical amongst those is the
lack of political will. Secondly, it is failure to translate party policy into government policy.
Thirdly is failure to translate policy into programmes and projects. But it is also about the
reluctance to avail resources, i.e. funding, time and personnel to devote to implementation
of policies. Policy may also be implemented through strategies and charters. However, these
strategies and charters must not seek to revise policy through a sleight of hand. In addition,
policy implementation may require the introduction of new legislation or amendment of
existing legislation as well as new institutional arrangements.
There is still extensive body of theory and concepts that have admittedly not been discussed
in this piece, such as meta problem, critical thinking, phenomenology and epistemology; not
because they are irrelevant or are of less importance, but as we have already stated, mainly
due to limited space and time.
So, as delegates for the ANC policy indaba converge, they need to appreciate that they carry
the hopes and aspirations of the masses of our country on their shoulders. It would be
better if they also understood the intricacies involved in a policy making process and how
choices are made. But they must also appreciate that their role is only the first phase of
policy formulation and that much wider processes will ensue after the conference.
We wish them luck, vision and wisdom!
Selepe is a student of Governance
and Social Change. He writes in his