Public Policy II: Institutions of Policy-Making; Types of Policy Analysis; Implementation, Outcomes and Evaluation; Socio-Economic, Political, Institutional and Cultural Constraints; Media, Public Opinion, Civil Society and Pressure Groups
2. Types of policy analysis
Policy analysis is the use of reason and evidence to choose the best policy among alternatives, or to understand what policies do. William Dunn describes it as producing information about policy problems, expected outcomes, preferred policies, observed outcomes and policy performance. The syllabus names three pairs of types, and each pair is defined by a different question.
| Type | Question it asks | Example |
|---|---|---|
| Empirical | What are the facts? What causes what? (the "is") | Did a school-meals scheme raise attendance? |
| Normative | What is desirable or right? (the "ought") | Should the scheme be universal or targeted? |
| Retrospective (ex post) | What happened after the policy was adopted? | Evaluation of a completed programme |
| Prospective (ex ante) | What will happen if a policy is adopted? | Forecasting the costs and effects of a proposed law |
| Descriptive | What is the policy and how was it made? (explaining without recommending) | An account of how a law passed Parliament |
| Prescriptive | What should be done? (recommending a course of action) | A committee report recommending reforms |
3. Policy implementation, outcomes and evaluation
Implementation is what happens between a policy's adoption and its results. The field was founded by Jeffrey Pressman and Aaron Wildavsky's Implementation (1973), whose long subtitle — How Great Expectations in Washington Are Dashed in Oakland — told the story of a federal job-creation programme that failed because so many decision points and clearances by many agencies were needed that even small probabilities of delay multiplied into failure (the "complexity of joint action"). Two schools followed.
| Aspect | Top-down | Bottom-up |
|---|---|---|
| Starting point | The policy decision and its objectives, set by central policy-makers | The front-line workers and the local network of actors who deliver the service |
| Key authors | Pressman and Wildavsky; Van Meter and Van Horn; Mazmanian and Sabatier (conditions for effective implementation); Hogwood and Gunn (conditions for "perfect implementation", 1984) | Michael Lipsky, Street-Level Bureaucracy (1980); Richard Elmore's "backward mapping" (1979–80); Benny Hjern |
| Implementation gap explained by | Unclear objectives, too many agencies, inadequate resources, weak control | The discretion of street-level workers, who in effect make policy as they ration scarce time and resources |
| Prescription | Clear goals, adequate resources, simple chains of command, compliance | Design policy backwards from the point of delivery; empower local actors |
Lipsky's street-level bureaucrats — teachers, police constables, health workers, welfare officials, and in India the patwari, the anganwadi worker and the village-level worker — work with inadequate resources, ambiguous goals and heavy caseloads, so they develop routines and simplifications that ration services and, in aggregate, become the policy citizens experience. Paul Sabatier's later advocacy coalition framework tried to synthesise the two schools. Outputs (what government produces — schools built, cards issued), outcomes (the changes in the target population — literacy rates, nutrition) and impacts (the longer-term, wider effects) must be distinguished; a programme can achieve its outputs and still fail on outcomes.
Evaluation assesses whether a policy achieved its objectives, at what cost, and why. It may be formative (during implementation, to improve it) or summative (at the end, to judge it); it may examine process or impact; and its methods range from cost–benefit and cost-effectiveness analysis to randomised controlled trials, quasi-experiments and participatory methods such as social audit. In India, the Programme Evaluation Organisation was set up under the Planning Commission in 1952 and is now part of NITI Aayog's Development Monitoring and Evaluation Office (2015); the CAG's performance audits and parliamentary committees add external evaluation.
4. Constraints on public policy
| Constraint | How it limits policy |
|---|---|
| Socio-economic | Scarce resources and fiscal limits; poverty and inequality; a large informal economy that is hard to reach; low literacy that limits awareness of entitlements |
| Political | Electoral cycles and short time-horizons; coalition compulsions; vote-bank politics; opposition from powerful interests; federal bargaining between Union and states |
| Institutional | Weak administrative capacity; fragmented departments and poor coordination; rigid rules; judicial review; weak local governments; the limits of data |
| Cultural | Traditions, religious beliefs, caste and gender norms that resist change — social-reform laws against dowry, child marriage or manual scavenging depend on changing attitudes as much as on enforcement |
5. Media, public opinion, civil society and pressure groups
The media shape policy chiefly through agenda-setting — Maxwell McCombs and Donald Shaw's study of the 1968 US presidential election (1972) showed that the media do not tell people what to think but what to think about — and through framing, which defines how an issue is understood. Investigative journalism exposes failures and scandals that force policy change; social media now allow rapid mobilisation but also spread misinformation. Public opinion sets the boundaries of what governments can do and is expressed through elections, surveys, protests and petitions; V. O. Key described it as the opinions held by private persons that governments find it prudent to heed. Civil society organisations contribute expertise, represent the unrepresented, run pilots that governments later adopt, and campaign — the Right to Information, the rural employment guarantee and the Right to Education all owed much to civil-society campaigns.
Pressure groups are organised groups that seek to influence policy without seeking to form a government — which distinguishes them from political parties. Gabriel Almond classified interest groups as associational (specialised, formally organised groups such as trade unions, business associations and professional bodies — in India, FICCI, CII, ASSOCHAM, the trade union federations), institutional (groups inside formal institutions such as the bureaucracy, the armed forces or churches that act as interest groups), non-associational (informal groups based on kinship, caste, religion, region or ethnicity) and anomic (spontaneous, often violent outbursts such as riots and demonstrations). Their methods include lobbying, representations, advertising and campaigns, litigation, and protest. Pluralist theory sees their competition as the essence of democratic policy-making; critics note that business and organised interests have far greater resources than the poor, and warn of regulatory capture.
Key takeaways
- Official policy-makers are the legislature, executive, bureaucracy and judiciary (which makes policy by direction in PIL); unofficial participants include parties, groups, media and civil society.
- Policy analysis is empirical or normative (facts or values), retrospective or prospective (after or before), descriptive or prescriptive (explain or recommend).
- Pressman and Wildavsky (1973) founded implementation studies; top-down theorists seek clear goals and control, while Lipsky (1980) and Elmore start from street-level discretion and backward mapping.
- Outputs, outcomes and impacts differ; evaluation may be formative or summative; India's Programme Evaluation Organisation (1952) is now in NITI Aayog's DMEO.
- The media set the agenda (McCombs and Shaw, 1972); Almond's interest groups are associational, institutional, non-associational and anomic; Kingdon's three streams open policy windows.
Practice questions (10)
Attempt each one before opening the answer. Every explanation names the tempting wrong option as well as the right one, because that is where marks are lost.
Street-Level Bureaucracy (1980), which argued that front-line workers in effect make policy through their discretion, was written by:
Show answer
Answer: C — Michael Lipsky
Lipsky founded the bottom-up view of implementation. Wildavsky co-wrote Implementation (1973), Sabatier developed the advocacy coalition framework, and Kingdon the multiple-streams model.An analysis carried out before a policy is adopted, forecasting its likely costs and effects, is:
Show answer
Answer: B — Prospective (ex ante)
Prospective analysis looks forward from before the decision. Retrospective analysis and summative evaluation look back after the fact; descriptive analysis explains without recommending.Which of the following are categories of interest groups in Gabriel Almond's classification? Select all that apply.
Show answer
Answer: A — Associational; B — Anomic; C — Non-associational
Almond's four are associational, institutional, non-associational and anomic. "Prismatic" belongs to Riggs's typology of societies.Assertion (A): In Pressman and Wildavsky's Oakland study, a well-funded federal programme largely failed to achieve its goals. Reason (R): The large number of decision points and clearances required by many agencies multiplied the chances of delay and breakdown. Choose the correct option.
Show answer
Answer: A — Both A and R are true, and R is the correct explanation of A
Both are true and R explains A: the "complexity of joint action" — many participants and clearances — meant that even individually likely agreements compounded into a low probability of success.McCombs and Shaw's agenda-setting research showed that the media chiefly influence:
Show answer
Answer: D — What issues people think about, rather than what they think about them
Their 1972 study of the 1968 election found that the salience of issues in the media matched the salience of issues for voters: the media tell people what to think about, not what to think.Which of the following correctly distinguish outputs, outcomes and impacts? Select all that apply.
Show answer
Answer: B — The number of anganwadi centres opened is an output; C — A fall in child malnutrition in the target area is an outcome
Centres opened are outputs and reduced malnutrition is an outcome. Meeting output targets does not guarantee outcomes, and spending is an input, not a measure of impact.Richard Elmore's "backward mapping" approach to implementation begins with:
Show answer
Answer: A — The behaviour to be changed at the point of delivery
Backward mapping starts at the last stage — the point where the policy meets the citizen — and works back up to decide what each level must do; forward mapping starts from the top-level decision.Which of the following is an example of judicial policy-making through public interest litigation?
Show answer
Answer: A — The Vishaka guidelines on sexual harassment at the workplace (1997)
In Vishaka the Supreme Court laid down binding guidelines in the absence of legislation, which governed until the 2013 Act. The other options are legislative and executive policy-making.In which year was the Programme Evaluation Organisation set up under the Planning Commission? Type a number.
Numerical answer — type the value.
Show answer
Answer: 1952
The PEO was set up in 1952 to evaluate community development programmes; it was later merged into NITI Aayog's Development Monitoring and Evaluation Office, created in 2015.In Kingdon's multiple-streams model, a "policy window" opens when:
Show answer
Answer: C — The problem, policy and politics streams come together
Kingdon argued that the three streams usually run independently; when they couple — a recognised problem, an available solution and a favourable political climate — a window opens for policy entrepreneurs to act.