.1 What This Unit Is About
Policy formulation decides what government should do; implementation determines what actually happens once that decision is made — the gap between the two is a core, recurring governance problem this unit dissects in unusual depth, naming eight distinct specific causes of implementation failure rather than treating “implementation problems” as one undifferentiated category.
.2 The Policy-Making Process: Six Stages
A useful learning sequence: problem diagnosis → agenda-setting/prioritization → option formulation → decision/adoption → implementation → evaluation/feedback. The syllabus’s important methodological caution, worth stating in any answer using this model: different textbooks divide the stages somewhat differently, and real policy processes are frequently non-linear — problems can re-enter the agenda after apparent resolution, implementation experience can reshape the original design, and evaluation findings can trigger entirely new formulation, rather than the process running neatly through six stages exactly once. State explicitly which model is being used rather than presenting any one sequence as a universally fixed law of how policy actually happens.
.3 Diagnosis and Cautions at Each Stage
Good diagnosis at any stage separates observed symptoms from underlying causes (a principle already emphasized repeatedly in the Public Administration chapter), identifies the relevant stakeholders, checks the actual evidence rather than assumption, and anticipates implementation constraints before a policy is adopted rather than discovering them afterward. Each stage specifically requires attention to legality, cost, distributional impact (who actually bears the costs and receives the benefits), feasibility, and potential unintended consequences.
.4 Eight Specific Diagnosed Causes of Implementation Failure
This unit’s most distinctive contribution is naming eight specific, analytically distinct causes of implementation failure — a genuinely useful diagnostic checklist for any implementation scenario:
- Inadequate training — implementation requires officials to genuinely understand new rules, systems, and skills; training must be connected to actual job requirements and reinforced by ongoing supervision, aligned incentives, and adequate resources, or the training itself becomes merely symbolic.
- Pervasive institutional inertia — arising from routines that reduce uncertainty for staff, genuine sunk investments in existing systems, risk-averse decision-makers, and incentive structures that reward continued compliance with existing practice rather than change. Breaking inertia requires visible leadership commitment, realigned incentives, genuine staff participation in designing the change, pilot testing before full rollout, and allowing gradual institutional learning rather than attempting an abrupt, complete transition.
- Lack of decision-making / a culture of delay — arising from unclear decision authority, fear of audit or blame for a wrong call, excessive approval layers, inadequate information to decide confidently, or genuine political uncertainty. The critical diagnostic point: a remedy must target the actual specific cause of delay — merely imposing a deadline on a process delayed by unclear authority or fear of blame will not resolve the underlying problem, since the deadline does nothing to clarify authority or remove the fear driving the delay.
- Lack of resources — resources for implementation include not only money but staff numbers, skills, technology, data, time, legal authority, and managerial attention; the frequently tested point is that a policy can be fully funded financially and still lack genuine implementation capacity if these other resource types (skilled staff, adequate data systems, sufficient managerial attention) are missing.
- Political interference and the ability to say “no” — professional administration requires genuine responsiveness to lawful political direction, combined with the institutional and personal capacity to resist instructions that are illegal, procedurally improper, or outside delegated authority. Institutional clarity, written procedures, and professional ethics (recall the Estacode discussion in Unit VI) are central to sustaining this balance in practice.
- Lack of research and policy analysis — weak underlying data or analysis leads directly to poorly designed policy from the outset; improvement requires credible data systems, genuine research capacity, systematic evaluation, real consultation, and transparent statement of assumptions (recall the good-policy-analysis standard from the Public Administration chapter’s Unit III).
- Lack of coordination among departments — cross-cutting policy issues require a clearly identified lead responsible actor, genuinely shared information, compatible timelines across the involved departments, explicitly agreed decision rights, and a working dispute-resolution mechanism. The frequently tested caution: simply creating a coordinating committee does not, by itself, guarantee actual coordination — without these specific supporting elements, a coordination committee can itself become another symbolic mechanism (recall Unit V’s symbolic-versus-substantive distinction) rather than a genuinely functioning one.
- Secrecy versus openness — some information legitimately requires confidentiality (genuine privacy, security, or ongoing-process concerns), but unnecessary secrecy weakens both institutional learning and accountability. The analytical question this unit asks is specifically whether a given instance of secrecy is legally justified and proportionate, not whether all government information should categorically always be public.
.5 Policy Evaluation
Policy evaluation assesses relevance (is the policy still addressing a genuine, current problem), effectiveness, efficiency, equity, sustainability, and unintended effects, as appropriate to the specific policy — with process evaluation (examining how implementation actually unfolded), outcome evaluation (examining resulting changes), and impact evaluation (examining attribution of change specifically to the policy) kept analytically distinct, directly paralleling the monitoring/evaluation and results-chain vocabulary already developed fully in the Public Administration chapter’s Unit III. Worked illustration: if a training program delivered 1,000 training sessions, that figure is an output; whether participants actually acquired the intended skills and subsequently used them at work is a genuinely separate outcome question, and one an output count alone cannot answer.
.6 Success and Failure of Government
Government performance should be judged against explicit stated objectives, actual evidence, and the surrounding context — the same evidentiary standard for governance-failure diagnosis already established in Unit I above, now applied specifically to evaluating overall government performance. The syllabus’s repeated, explicit caution: avoid simplistic “success” or “failure” labels when outcomes genuinely depend on multiple actors (other levels of government, external partners) or on external shocks (economic conditions, natural disasters) beyond any single government’s control — attributing a complex, multi-causal outcome entirely to “government success” or “government failure” oversimplifies exactly the kind of attribution problem already flagged for impact evaluation above.
Key takeaways for Unit VII
- The eight specific implementation-failure causes named in this unit (training, inertia, delay, resources, political interference, weak analysis, poor coordination, secrecy) form a genuinely useful diagnostic checklist — a strong implementation-gap answer identifies which specific one(s) from this list are operating in a described scenario, rather than describing “implementation problems” vaguely.
- The remedy-must-match-cause principle (illustrated with the delay example) applies across all eight causes: a generic remedy (more funding, a new committee, a stricter deadline) that does not address the actual underlying cause will not resolve the implementation failure.
- Policy evaluation’s process/outcome/impact distinction, and the caution against oversimplified success/failure labels for multi-causal outcomes, both directly extend the Public Administration chapter’s evaluation and governance-failure frameworks into this subject’s specific policy-implementation context.