The Role of the Government in Decision Making
Chapter Five
Syllabus topic 1.1.2, "Exercise of political power and role of the Government in decision making"
Pages 19 to 22 of 506
In one line
A government decision is not a moment when somebody chooses; it is a process with stages, and the models of decision making are competing accounts of what actually happens in that process.
In the wording a student can write in an exam: the government's role in decision making is to convert demands arising in society into binding public decisions, through a policy cycle of agenda setting, formulation, adoption, implementation and evaluation; and how far that process is genuinely rational is contested between the rational, incremental, bureaucratic, garbage-can and mixed-scanning models.
The policy cycle
The stages are worth learning in order, because a question about the government's role can be answered by walking down them.
1. Agenda setting. Out of the thousands of problems in a society, a few become things the government is going to do something about. This is the stage the second face of power operates on, and it is decided as much by what is kept off the list as by what goes on it.
2. Formulation. Options are worked out, usually by officials, sometimes by an expert committee, a commission or a law commission, and increasingly by consultants.
3. Adoption. The choice is made and given legal form: a statute, an ordinance, a notification, a scheme, a budget line.
4. Implementation. The decision is carried out by an administrative machine that has its own interests, capacities and shortages. Most policy failures are failures here.
5. Evaluation. What happened is measured against what was intended, by the audit institution, by parliamentary committees, by courts, by the press and by researchers.
6. Feedback. Evaluation changes the agenda, and the cycle starts again.
The models, and what each one claims
The rational model
Also called the synoptic or comprehensive model. The decision maker defines the goal, lists all the alternatives, works out the consequences of each, and picks the one that best achieves the goal at least cost.
Its virtue. It is the standard against which everything else is measured, and it is the model a court applies when it asks whether a decision was reasonable and whether relevant considerations were taken into account.
Its problem. No government can do this. Nobody knows all the alternatives, the consequences cannot be fully calculated, goals conflict with each other, and time and money run out. Herbert Simon called the result bounded rationality: decision makers work within limits of information, capacity and time, and therefore do not maximise but satisfice, taking the first option that is good enough rather than searching for the best.
The incremental model
Charles Lindblom's answer, and its nickname is the whole argument: the science of muddling through. Governments do not start from first principles. They start from what is already being done and change it a little. Last year's budget plus five per cent. The existing scheme with an extra category added.
The Role of the Government in Decision Making
Why it happens. It is cheap, it is politically safe, mistakes are small and reversible, and it does not require agreement on ultimate goals, only on the next step.
Its criticisms. It is conservative, it entrenches whoever benefits from the status quo, and it cannot produce the large change a real crisis needs. A country that had reformed its economy incrementally would not have had 1991.
The bureaucratic and organisational model
Graham Allison's contribution. On this account a government decision is not the choice of a single mind at all. It is the output of organisations following their standing procedures, and the resultant of bargaining between departments each of which has its own view, its own budget and its own turf. Ask what a government decided and you often find that no one person decided it.
Its Indian application is obvious. A decision that needs Finance, Home, Law and a State government to agree will look nothing like what any of them would have done alone.
The garbage-can model
Cohen, March and Olsen described organisations in which problems, solutions, participants and choice opportunities are thrown together and connected more or less by chance: a solution that has been sitting in a drawer meets a problem it was not designed for, and gets adopted because both happened to be present when a decision had to be made. Governments produce more decisions of this kind than they admit.
Mixed scanning
Amitai Etzioni's compromise, and the one to quote when a question asks you to choose. Big decisions are taken with a broad, rational scan of the whole field, and the many small decisions that follow are taken incrementally within the frame the big one set. This describes real government better than either pure model.
Who actually decides in India
The Council of Ministers is where a major decision is formally taken, on a note prepared by a department, and its decisions are collective.
The department frames the note, and framing is power: the options that reach the meeting are the options the department wrote down.
The Cabinet Secretariat and the Finance Ministry are the two gatekeepers almost nothing gets past.
Expert bodies produce the material that shapes the option set: commissions, law commissions, and the policy think tank that replaced the Planning Commission.
Parliament adopts what needs a statute, scrutinises through committees, and controls money.
The courts review, and in India also sometimes direct.
The States implement most of what touches an ordinary person's life, because health, police, land, agriculture and local government are State subjects.
The Role of the Government in Decision Making
A worked example
A State decides to give a monthly cash transfer to women heads of poor households.
Agenda setting. The demand exists for years and goes nowhere. It reaches the agenda when an election is a year away and a neighbouring State's scheme is reported to have worked. Notice that the merits did not change; the political opportunity did.
Formulation. The Women and Child Development Department prepares a note with three options: a universal transfer, a transfer to households below the poverty line, and a transfer with an income cut-off and an exclusion list. Finance costs each. The department's preference is visible in the order the options are listed.
Adoption. The Council of Ministers approves the third option. This is the moment the model matters: it looks like a rational choice among three, but the three were chosen by officials, and a fourth option, spending the same money on childcare, was never written down.
Implementation. The transfer requires bank accounts, an eligibility list, a grievance mechanism and a helpline. The list is built from an old database, so a large number of eligible women are missing and a smaller number of ineligible ones are included. This is where the scheme is actually decided, by clerks applying a proforma.
Evaluation. The audit institution reports on exclusion errors. A newspaper finds women who were left out. A petition is filed. A committee recommends changes.
Feedback. Next year's version raises the cut-off and adds an appeal. That is incrementalism, and it will now continue for a decade.
Distinctions that carry marks
| Rational model | Incremental model | |
|---|---|---|
| Starting point | The goal, defined afresh | What is already being done |
| Range of options | All of them | Those marginally different from the present |
| Test applied | Which best achieves the goal | Which can be agreed on now |
| Assumes | Full information and settled goals | Neither |
| Strength | A standard of good decision making | Describes what governments really do |
| Weakness | Impossible in practice | Conservative, and helpless in a crisis |
| Associated with | The classical account; Simon's critique | Lindblom, "the science of muddling through" |
What this is NOT
Decision making is not the same as policy. A policy is a settled line of action; a decision is one act within it. Governments take thousands of decisions inside a single policy.
A decision is not made where it is announced. It is usually made in the drafting of the note that limited the options.
Not deciding is a decision. A government that lets a file sit has decided, and the decision binds exactly as much as an order would.
Implementation is not a technical afterthought. Most schemes that fail are well designed and badly delivered, which is why evaluation studies concentrate on delivery.
The Role of the Government in Decision Making
Quick revision
- The policy cycle: agenda setting, formulation, adoption, implementation, evaluation, feedback.
- Rational model: define the goal, list all options, compute consequences, maximise. Impossible; Simon's bounded rationality and satisficing.
- Incremental model, Lindblom: start from the present and change it a little. "Muddling through." Safe and conservative.
- Bureaucratic and organisational model, Allison: the output of standard procedures and inter-departmental bargaining, not the choice of one mind.
- Garbage-can model: problems and ready-made solutions meet by chance.
- Mixed scanning, Etzioni: broad scanning for the big decision, incrementalism for the rest. The best single answer.
- In India: the department frames, the Council of Ministers adopts, Finance and the Cabinet Secretariat gatekeep, Parliament legislates and scrutinises, the courts review, and the States implement.
Test yourself
1. Set out the stages of the policy cycle. Agenda setting, in which a problem becomes one the government will act on; formulation, in which options are worked out, usually by officials or an expert body; adoption, in which the choice is made and given legal form; implementation, in which the administration carries it out; evaluation, in which outcomes are measured against intentions; and feedback, in which the evaluation reshapes the agenda and the cycle begins again.
2. What is bounded rationality, and whose idea is it? Herbert Simon's. Decision makers cannot achieve the comprehensive rationality the classical model demands, because information is incomplete, consequences cannot be fully computed, goals conflict, and time and money are limited. They therefore work within those bounds and satisfice, adopting the first alternative that is good enough rather than searching for the best available one.
3. Explain the incremental model and state two criticisms of it. Lindblom's model holds that governments begin from what is already being done and change it marginally, because that is cheap, politically safe, and needs agreement only on the next step rather than on ultimate goals. It is criticised as conservative, since it entrenches whoever benefits from the present arrangement, and as inadequate to a crisis, since a situation that requires large and rapid change cannot be met by successive small ones.
4. How does Allison's bureaucratic model change the question "what did the government decide?" It denies that there was a single deciding mind. On this account a government decision is the output of organisations executing their standard operating procedures, and the resultant of bargaining among departments with different interests, budgets and jurisdictions. The right question becomes which organisations were involved, what their routines produced, and how the bargain between them was struck.
5. What is mixed scanning, and why is it the most realistic model? Etzioni's proposal that decision making operates at two levels: fundamental decisions are taken after a broad but not exhaustive scan of the whole field, and the many small decisions that follow are taken incrementally within the frame those fundamental ones set. It is realistic because it accounts for both the rare large reform and the ordinary daily adjustment, which neither the rational nor the incremental model can do alone.
The rest of this subject
These notes are cut from the University's printed syllabus. Open the syllabus itself, or the past papers, for the same subject.