By the end of this chapter you'll be able to…

  • 1Distinguish accountability, control and responsibility, and place the Friedrich-Finer debate correctly
  • 2Compare legislative, judicial, executive and citizen controls by what each can reach
  • 3State Dicey's position and the critique of it, and treat administrative law as the rule of law applied to administrative power
  • 4Give delegated legislation with its necessity, its dangers and the three families of control
  • 5Apply Riggs's formalism, heterogeneity and overlapping to named Indian divergences
  • 6Trace the movement of the development concept and hold both findings on bureaucracy and development
  • 7Locate a policy failure at its stage before prescribing a remedy
  • 8Give each budget form with its failure mode and each cycle stage with its control point
💡
Why this chapter matters in UPSC CSE
This is the applied half of the theory paper, and its unifying question is how administration is held to account and improved. Every mechanism here is scored on what it can and cannot reach, so the examinable content is comparison and limit rather than description.

Accountability, Law, Policy & Financial Administration

Weightage: Paper I, Units 5–12 — the applied half of the theory paper. Its unifying question is how administration is held to account and improved, and every unit here supplies a different instrument for doing it.

1. Accountability and control

The concepts first, because they are not the same thing. Accountability is the obligation to answer for conduct to someone with a right to require the answer — it is retrospective, relational and includes the possibility of consequence. Control is the capacity to direct or restrain conduct — it operates prospectively and continuously. An official may be controlled without being accountable, as under a chain of command with no external answerability, and accountable without being controlled, as with a regulator answering to the legislature after the fact.

Responsibility adds a third element: the internalised obligation to act rightly, which operates where neither control nor accountability reaches. This is why the Friedrich–Finer debate is the unit's organising controversy.

Finer's position: administrative responsibility must be secured by external control — legislative, judicial and hierarchical — because the servants of the public are not to decide their own course, and reliance on the official's conscience substitutes what the official thinks is right for what the public has authorised. His argument is that only enforceable answerability can be relied upon.

Friedrich's position: modern administration is too technical, too fast and too discretionary for external control to reach the decisions that matter. Responsibility must therefore be secured internally, through professional standards and responsiveness to popular sentiment and to technical knowledge — the "dual standard" of technical knowledge and popular sentiment.

The resolution, which is what an answer should give: both are right about the limits of the other's mechanism. External control is indispensable and insufficient — it reaches legality, procedure and gross failure, and it cannot reach the exercise of discretion in the ordinary case, which is where most administrative harm occurs. Internal responsibility is necessary and unverifiable — it operates precisely where external control cannot, and there is no way to confirm it is operating. Practical systems combine them, and the design question is where the boundary lies, not which is correct.

The three external controls, compared by reach

This comparison, rather than a description of each, is what accountability questions are asking for.

Legislative control operates through questions and supplementaries, which reach specific administrative acts; debates and motions, which reach policy; the budget, which reaches allocation before the fact; the financial committees — public accounts, estimates and public undertakings — which reach expenditure after audit; and committees on subordinate legislation and on assurances, which reach delegated rule-making and undertakings given in the house.

Its limits: it operates after the fact; it depends on parliamentary time, which is scarce and shrinking; it depends on the opposition's strength, so it is weakest exactly when the executive is strongest; the financial committees examine a small fraction of transactions and report long after; and party discipline means the legislature's majority has limited incentive to press its own government.

Judicial control operates through judicial review of administrative action on grounds of illegality — acting beyond power or under a misdirection of law; irrationality — a decision so unreasonable that no reasonable authority could reach it; and procedural impropriety — breach of natural justice or of the procedure prescribed. It also operates through the writs, through statutory appeals, and through the expanded standing that permits action by those not personally aggrieved.

Its limits: it reaches legality and procedure, not the merits of a policy or the quality of implementation; it is expensive and slow, so it is available disproportionately to those with resources; it is episodic, acting only where invoked; and its expansion into policy raises the objection that unelected judges are substituting their judgement for the elected government's.

Executive control operates through the hierarchy and the power of direction; financial control through budget allocation, sanction and re-appropriation limits; personnel control through appointment, transfer, promotion and discipline; rule-making and the issue of instructions; and inspection and internal audit.

Its limits: it is control of the administration by the administration, so it fails precisely where the failure is systemic or where superiors share the interest of subordinates; and its most-used instrument — transfer — is simultaneously a control device and, when politically directed, a means of undermining independent conduct.

The general finding: each mechanism reaches something the others cannot, none reaches the point of delivery where the citizen actually encounters administration, and that gap is what the citizen-facing instruments were created to fill.

Citizen and administration

Citizen's charters: published statements of the standards of service a citizen may expect, the time limits applicable, and the remedy for failure. Their weakness, which the Indian experience demonstrates, is that most are not enforceable — a charter that states a standard without a consequence for breach is a declaration of intent, and the reform proposal that follows is to make the commitments justiciable and to attach compensation.

Right to information: the strongest of the citizen instruments, and the reason is structural. It creates an individually claimable right, enforceable through a designated officer, an appellate authority and a commission with the power to impose penalties on the official who withholds. It therefore does not depend on any administrator choosing to comply, which is why it has been implemented where instruments requiring administrative forbearance have not. Its limits: exemptions, particularly on file notings and internal deliberation where these are restricted; delays in the appellate chain; the difficulty of using it where the applicant is not literate or is vulnerable to retaliation; and its confinement to information, which reveals a failure without correcting it.

Social audit: verification of a programme's records against the account given by its beneficiaries, conducted in a public assembly. Its mechanism is that it substitutes the beneficiary's testimony for the record where the two diverge, which reaches falsification that documentary audit cannot. Its conditions are the ones that make it work or fail: independence of the auditing body from the implementing agency, mandatory public hearings, and protection for those who testify.

Grievance redress: internal complaint machinery, ombudsman-type institutions, and statutory service guarantees at state level. The design question is whether the mechanism has the power to direct a remedy or only to recommend one, and the second is much weaker.

Media, interest groups, civil society and voluntary organisations: these exercise influence rather than control, working by exposure, by mobilising opinion, and by litigation. Their contribution is reach — they detect what the formal mechanisms miss and place items on the agenda. Their limits are that they are selective, attending to what is visible and newsworthy; unaccountable themselves, with no mandate from those they claim to represent; and unevenly distributed, being weakest where administration is worst.

2. Administrative law

Meaning: the body of law governing the organisation, powers, duties and procedures of administrative authorities, and the remedies available against them. It is concerned with the exercise of public power — how it is conferred, how it must be exercised, and how it is controlled.

Its significance: the modern state legislates through administration, adjudicates through tribunals, and touches citizens principally through officials exercising discretion. Administrative law is the discipline that subjects that power to legal standards, and its growth is a direct function of the growth of the administrative state.

Dicey's position, which is the unit's set-piece. Dicey held that England had no administrative law and should have none, on the ground that the rule of law required three things: the absence of arbitrary power, so that no one is punished except for a breach of law established before an ordinary court; equality before the law, so that officials are subject to the same law and the same courts as citizens; and a constitution that is the result of ordinary law rather than its source. He contrasted this with the French droit administratif, with its separate body of law and separate tribunals for administrative disputes, and treated it as a system of official privilege incompatible with equality before the law.

The critique of Dicey, which the answer must give. He misunderstood droit administratif, which developed protections for the citizen against the administration at least as strong as the English common law then provided. He ignored the discretionary powers already conferred on English officials by statute, so his description was inaccurate even of his own system. And his position could not survive the growth of the welfare and regulatory state, which required conferral of discretion on a scale his framework had no place for. The modern position is that administrative law is not the negation of the rule of law but its application to administrative power — the rule of law survives as the requirement that discretion be conferred by law, exercised for the purpose for which it was given, and subject to review.

Delegated legislation: rules, regulations, by-laws and orders made by the executive under authority conferred by a statute.

Why it is unavoidable: legislative time is scarce; the subject matter is technical and requires expertise the legislature lacks; the detail must be adjustable without returning to the house; emergencies require speed; and experimentation is impossible if every detail is fixed in primary law.

The dangers: excessive delegation, where the legislature transfers its essential function rather than filling in detail; the delegation of the power to modify the parent statute itself, which is the most objectionable form; sub-delegation beyond what was authorised; retrospective operation; and the sheer volume, which defeats scrutiny.

The controls: legislative control, through the requirement of laying before the house, affirmative or negative resolution procedures, and scrutiny by a committee on subordinate legislation; judicial control, on the grounds that the parent statute is itself unconstitutional, that the rule exceeds the power conferred, that it is procedurally defective, or that it is manifestly arbitrary; and procedural control, through requirements of prior publication, consultation with affected interests and consideration of objections. The essential legislative function doctrine is the constitutional limit: the legislature may delegate the working out of detail but must itself lay down the policy and the standard.

Administrative tribunals: bodies outside the ordinary court hierarchy that adjudicate disputes in a defined field.

The case for them: speed, since the ordinary courts are congested; expertise, since the adjudicator can be a specialist in the field; cheapness and informality, with relaxed procedure and evidence rules; and relief of the courts.

The objections: independence is weaker where members are appointed by, and dependent for reappointment on, the executive whose decisions they review; procedural informality can mean inadequate reasoning and uneven standards; exclusion of the ordinary courts removes a protection; and the multiplication of tribunals produces inconsistency. The settled position is that tribunals supplement rather than supplant judicial review, and that their independence in appointment and tenure is the condition of their legitimacy.

3. Comparative public administration

The founding claim is the ecological one: administrative systems are shaped by their historical, social, economic, political and cultural environment, so a structure that works in one setting will not behave the same way when transplanted. This was advanced against the assumption that administrative principles are universal, and it is the field's permanent contribution.

Why the field arose when it did: the post-war interest in development, the transfer of administrative models to newly independent states, and the observation that the transfers were not producing the predicted results.

Riggs's models, which are the examined content.

Agraria and industria: ideal types of the administrative systems of agricultural and industrial societies, contrasted on ascription against achievement, particularism against universalism, diffuse against specific roles, and limited against extensive mobility. The pair was quickly criticised as a dichotomy that left the actual cases in between unexplained, which is the problem the next model addresses.

The prismatic society, his central construct. The metaphor is a prism: fused light entering, diffracted light emerging, and the prismatic condition in between where the separation is partial. Traditional society is fused, with a single structure performing many functions; modern society is diffracted, with specialised structures for specialised functions; transitional society is prismatic, with new specialised structures formally established while the old fused ones continue to operate.

Its three features, which must be usable rather than nameable.

Formalism is the gap between the prescribed and the practised — between what the law says and what happens, between the official rate and the actual rate, between the organisation chart and the working arrangement. It is not simply non-compliance; it is a systematic divergence that both parties understand and rely on.

Heterogeneity is the simultaneous presence of very different systems within one polity — a modern secretariat and a customary local order, a formal credit market and a moneylender, industrial employment and subsistence cultivation — with the consequence that a single administrative rule encounters radically different conditions.

Overlapping is the intrusion of ascriptive, familial and communal considerations into formally rational structures, so that the formal organisation is penetrated by the social one. His sala is the prismatic administrative office, which is neither the fused chamber nor the diffracted bureau, and in which appointment, promotion and decision are influenced by kinship and community alongside the formal criteria.

The application to India should be made explicitly, since it is what the question is testing. Formalism explains the persistent gap between statutory entitlement and actual delivery, and why adding another rule rarely closes it. Heterogeneity explains why the same statute produces different administration in different districts. Overlapping explains why formal recruitment and posting rules coexist with informal influence, and why accountability mechanisms designed on the assumption of impersonal administration underperform.

The critique, which must be given.

The models are descriptive rather than predictive: they classify a condition without generating testable propositions about what will happen. The ecological approach explains everything after the fact and forecasts nothing, which is the standard objection to any framework that treats the whole environment as the cause. The categories are constructed from a Western developmental endpoint, so the prismatic society is defined by what it is not yet, which imports a teleology. The vocabulary is elaborate and largely unused outside the framework. And the treatment is static, describing a condition without explaining the movement through it.

The present status of the field: comparative public administration lost momentum after its initial period, principally because of the first objection — a framework that generates few testable propositions cannot sustain a research programme. What survives and has grown is comparative governance and public management research, which is empirical and indicator-based, comparing systems on measurable dimensions such as bureaucratic quality, corruption, and the design of civil service systems. The shift is from grand models to measured comparison, and the ecological insight survives inside it as the finding that reforms transfer badly across institutional contexts.

4. Development dynamics

The concept of development and its movement. The early formulation identified development with economic growth, measured by output per head. The successive broadenings are examinable: growth with distribution, adding the question of who receives it; basic needs, specifying food, shelter, health and education as the content; human development, defining development as the expansion of people's capabilities and choices rather than of output, with the associated composite measurement; and sustainable development, adding the intertemporal constraint. Each broadening was a response to an observed failure of the previous measure to track what was intended.

Development administration: the administration of development, and simultaneously the development of administration. Its early profile was the state as the agent of development, the administrator as mobiliser and planner, and administrative capacity as the binding constraint — with the prescription of capacity-building, planning machinery, and a development-oriented bureaucracy distinct from the law-and-order administration.

The changing profile followed the accumulated evidence: programmes delivered through existing local structures were captured by those structures; capacity was not the only constraint, since political will and incentive structures mattered at least as much; and top-down delivery did not produce the participation that sustained outcomes require. The emphasis shifted to decentralisation, participation, beneficiary involvement and, later, to rights-based approaches that convert a discretionary benefit into a claimable entitlement.

The anti-development thesis attacks the premise rather than the delivery. Its claims: development as conventionally defined is a discourse that constituted the societies it addressed as deficient; it displaced functioning arrangements — subsistence systems, commons, local knowledge — that were adequate to their setting; it imposed a single model derived from one historical experience as though it were universal; and it served the interests of the institutions and professionals who administer it.

Assessment: the thesis is strong as critique and weak as programme. It identifies real and unexamined assumptions, and it has no answer for populations who want the health services, education and infrastructure that development promises. The defensible use in an answer is to state it as an objection that forced the reconsideration of what development means, and then note that the reconsideration produced the capability and rights-based formulations rather than the abandonment of the project.

Bureaucracy and development is the unit's central question, and the honest answer holds two findings that pull in opposite directions. The Weberian features — merit recruitment, rule-boundedness, security of tenure, separation of office from private means — are positively associated with state capacity and with restraining corruption, and the successful late developers had strong meritocratic administrations. The same features produce rigidity, procedural delay and poor responsiveness in the delivery functions development requires.

The resolution in practice, which is the answer's payoff: administrations have retained the classical form for regulatory and rule-applying functions and sought other arrangements — decentralised bodies, autonomous societies, mission-mode structures, community delivery — for developmental ones. This is the mechanistic-organic distinction applied without being named, and it explains why the Indian administrative landscape contains both forms simultaneously.

Strong state versus market. The market case rests on public choice arguments about bureaucratic incentives, on the informational limits of central allocation, and on the record of state enterprise. The state case rests on public goods and externalities that markets under-provide, on distributional objectives markets do not pursue, on the coordination of investment where returns depend on complementary investments, and on the historical record of the successful late developers, all of which had strong states directing capital. The contemporary position is that the dichotomy is false: markets require state-supplied institutions — property rights, contract enforcement, regulation, competition policy — so the question is not state or market but the quality of the state's capacity to construct and regulate markets.

The impact of liberalisation on administration in developing countries is the concrete form of the shift. The state's role moves from producer to regulator, enabler and guarantor, requiring capacities the producer state never needed: regulatory expertise, contract design and management, and the ability to hold private providers to public purposes. It lengthens the accountability chain, since a service delivered by a contracted provider under a regulator is far less traceable than one delivered by a department. It exposes administration to constraints set outside the polity in trade rules, capital mobility and standards regimes. And it produces a characteristic failure where the regulatory capacity is created after the liberalisation rather than before it.

Women and development is a named unit and is answered through a shift in framing: from welfare, treating women as beneficiaries of provision; through equity and anti-poverty framings; to empowerment, treating women as agents with independent claims on resources and on decision-making. The corresponding analytical move is from women in development, which sought to integrate women into existing development programmes, to gender and development, which examines the relations between women and men and the structures producing the inequality — a change of object, not just of emphasis.

The self-help group movement is the examined case. Its mechanism: small groups of women save regularly, lend internally, and become creditworthy to a bank on the strength of group liability substituting for collateral, which is what makes lending to the assetless viable. Its documented effects: increased savings and access to institutional credit at rates far below the moneylender's; reduced dependence on informal credit; increased mobility and participation in household decisions; and a base for collective action in the village. Its limitations, which must be stated: the credit is small relative to what an enterprise needs; benefits concentrate among the less poor within the group, since the poorest cannot sustain regular savings; group formation is sometimes treated as a substitute for the wider entitlements it cannot deliver; and the model's success is heavily dependent on the quality of the promoting institution.

5. Personnel administration

Human resource development as the frame: the shift from personnel administration as the management of a establishment — recruitment, records, discipline — to the development of a workforce's capabilities over a career.

Recruitment: the choice between merit and patronage systems is settled in favour of merit; the live questions are the basis of merit and the structure of the service. Generalist against specialist is the enduring Indian debate: the generalist's case is that the higher administrator's task is coordination and judgement across fields, and that specialisation narrows perspective; the specialist's case is that modern administration is technical and that generalists make decisions they cannot evaluate. The contemporary resolution is a mixed structure — generalists for coordination and specialists for technical domains, with lateral entry as the mechanism for bringing in expertise mid-career, and with the objections to lateral entry being the effect on the career prospects of serving officers and the risk to the neutrality of appointment.

Training is examined for its types and its purposes: induction training, which imparts the role; in-service and refresher training, which updates; and specialised training for a functional assignment. Its recurring weaknesses in practice are that it is treated as a break rather than an investment, that it is not linked to posting or promotion, that content is dated relative to the work, and that evaluation of training's effect on performance is rarely attempted.

Position classification groups posts by duties and responsibilities rather than by the person occupying them, and is the basis of a rational pay structure. Its advantages are equal pay for equal work, clear career paths and a rational basis for recruitment and training. Its costs are rigidity, the administrative burden of continuous reclassification, and the difficulty of accommodating jobs that change.

Career advancement: the tension between seniority, which is objective, predictable and protective of independence, and merit, which rewards performance but requires a reliable appraisal system and is vulnerable to favouritism. The practical Indian position combines them, with seniority dominant at lower levels and merit weighted more heavily at senior ones.

Performance appraisal: the movement from confidential reporting to open, disclosed appraisal with numerical assessment and a right to representation; and further to performance agreements specifying outputs against which units are assessed. Its persistent weaknesses are inflation, in which almost all officers receive high ratings, which destroys the instrument's discriminating power; the halo effect and central tendency; the assessment of traits rather than results; and the absence of consequence attached to the outcome.

Discipline: the framework of conduct rules, the distinction between minor and major penalties, the requirement of a charge, inquiry and hearing before major penalties, and the constitutional protection of civil servants against dismissal or reduction in rank without a reasonable opportunity to be heard. The design tension is exact: the protection exists to secure independence from arbitrary political direction, and the same protection makes the removal of a genuinely non-performing or corrupt official slow and difficult. Every reform proposal here is a proposal about where on that trade-off to sit, and an answer should say so rather than treating delay as a mere inefficiency.

Employer-employee relations: staff associations and service unions, joint consultative machinery, and the question of the right to strike in essential public services — with the standard position that the compensating mechanism for restricting it must be effective and binding arbitration, failing which the restriction is one-sided.

Grievance redressal: internal channels, designated grievance officers, and appeal to statutory bodies. The design question is again whether the body can direct or only recommend.

Code of conduct and administrative ethics: rules prohibiting conduct unbecoming, requiring integrity and devotion to duty, regulating acceptance of gifts, outside employment, and the declaration of assets; and conflict-of-interest provisions including post-retirement employment restrictions. The analytical point: a code prohibits conduct and cannot supply the judgement that ethics requires, which is why codes are necessary and insufficient — they establish a floor and leave the substantive discretionary choices, which is where administrative harm actually occurs, to the official's own standards.

6. Public policy

Models of policy-making, each with its claim and its critique.

The rational model: comprehensive analysis of alternatives against stated objectives, selecting the maximising option. Critique: informational and cognitive limits make it unattainable; it assumes agreed and ranked objectives, which the political process does not supply.

The incremental model: policy proceeds by small departures from the status quo; means and ends are chosen together; agreement substitutes for demonstrated correctness; and analysis is drastically limited. Critique: conservative by construction, unable to address problems requiring discontinuity, and it ratifies the existing distribution of influence.

Mixed scanning: broad scanning for fundamental decisions, incremental detail within the frame they set. Critique: no criterion for identifying the fundamental decisions, which is the judgement it needs.

The institutional model: policy is what institutions produce, so its content is shaped by institutional structure — federalism, separation of powers, and the rules of legislative procedure — rather than by preferences alone.

Group theory: policy is the equilibrium reached in the struggle between interest groups, with the state as arena rather than actor. Critique: it under-weights the state's own interests and the asymmetry of organisation, since concentrated interests organise and diffuse ones do not.

Elite theory: policy reflects the preferences of a governing elite and flows downward, with the mass influencing it only marginally. Critique: it is difficult to falsify and understates the effect of electoral competition.

Systems theory: policy as the output of a political system converting inputs of demand and support, with feedback.

Game theory for interdependent decisions, and the garbage can for organised anarchies.

Policy streams: the model in which problem, policy and political streams flow independently and a policy is adopted when a window opens and an entrepreneur couples the three. This is the most usable for explaining timing — why a solution long available is adopted at a particular moment — which is the question most other models cannot answer.

The policy process and where each stage fails.

Agenda-setting: how a condition becomes a problem requiring action. The failure here is non-decision — the exclusion of issues from the agenda, which is a use of power invisible to any study of decisions taken.

Formulation: the design of alternatives. The failures are inadequate diagnosis, the selection of instruments unmatched to the cause, and the absence of an implementation assessment at the design stage.

Legitimation: securing authority through legislation, budgetary provision or executive decision.

Implementation: the stage the discipline neglected longest and where most policy fails. The identified causes are the number of clearances required, since a chain of independent approvals compounds delay multiplicatively; the gap between the policy's assumptions and field conditions; the discretion of street-level bureaucrats, whose coping behaviours under caseload pressure effectively remake the policy; the absence of matched resources; and goal ambiguity, where a policy passed by aggregating support carries incompatible objectives into implementation.

Monitoring and evaluation: the distinction is examinable — monitoring tracks inputs, activities and outputs against plan and is continuous; evaluation assesses outcomes and impact against objectives and is periodic. The persistent problems are that outcome data is scarce, that attribution to the intervention is difficult without a comparison group, that evaluation is conducted by the implementing agency, and that findings rarely feed the next cycle.

Review and termination: policies are far easier to start than to stop, because a constituency forms around any programme, and termination requires a decision against an organised interest in favour of a diffuse one.

State theories and policy formulation, a named syllabus item. The pluralist state is an arena in which competing groups produce policy as their resultant. The elite state produces policy reflecting a cohesive governing group. The Marxist state, in its instrumentalist form, acts for the dominant class, and in its relative autonomy form acts with independence from any particular capitalist interest in order to secure the conditions for accumulation as a whole. The institutionalist state has interests and capacities of its own, and policy reflects what its structure permits. The developmental state directs investment with a competent insulated bureaucracy and a coherent development project.

The use of these in an answer: each predicts a different explanation of the same policy, and naming which explanation a question's evidence supports is the analytical move. A policy favouring a concentrated industry over diffuse consumers is a group-theory outcome; a policy pursued against the preferences of every organised interest is evidence of state autonomy.

7. Techniques of administrative improvement

Organisation and methods: the systematic study of an organisation's structure, procedures and methods with the object of simplification — examining whether each step is necessary, whether forms and returns are used, and whether the sequence can be shortened. Its recurring finding is that a large share of steps exist because they once served a purpose that has lapsed.

Work study and work management: method study, which establishes the best way to do a task, and work measurement, which establishes how long it should take, together supplying the basis for staffing norms and workload assessment. Its limitation in administration is that it applies well to repetitive countable work and badly to judgemental work, and applying it to the latter measures the measurable and neglects the rest.

Network techniques: PERT and CPM represent a project as a network of activities with dependencies, identify the critical path — the longest sequence of dependent activities, which determines the project's minimum duration — and show where delay in one activity delays the whole project and where slack exists. CPM uses deterministic time estimates, PERT uses probabilistic ones with optimistic, most likely and pessimistic estimates. Their administrative value is that they identify which delays matter, which is the question a project review otherwise cannot answer. Their limits: they require an accurate activity list and reliable estimates, and they model the schedule and not the political and clearance dependencies that actually delay public projects.

Management information systems: the structured collection, processing and reporting of information for decision-making. The design principle that carries marks: an MIS should be built around the decisions it supports, not around the data that happens to be collected — and the standard failure is a system that generates volumes of reports nobody uses while the information a decision requires is unavailable.

E-governance and information technology is the most examined item in the unit. Its models are conventionally given as G2C, G2B, G2G and G2E, and its stages as presence, interaction, transaction and transformation, with the last requiring changes to process rather than only to interface.

What it achieves: reduced discretion at the point of transaction, which removes the occasion for rent-seeking — the mechanism by which computerisation of land records, registration and permits reduces corruption; transparency, since a digital record is harder to alter silently; traceability, so that delay is attributable to a step and an officer; reach, through service centres and mobile channels; and process integration across departments.

What it does not achieve, and the honest assessment: computerising a process without redesigning it encodes the existing complexity, so the number of steps is preserved and only the medium changes. The digital divide — connectivity, device access, literacy and language — excludes exactly the population that most needs the service, and a service available only digitally is a service withdrawn from them. Exclusion errors in authentication and beneficiary matching deny entitlements to people who hold them, and the harm is concentrated among the least able to appeal. Data protection and surveillance concerns follow from the aggregation of personal data. And back-end process change is the hard part, which is why the transformation stage is reached far less often than the transaction stage.

8. Financial administration

Monetary and fiscal policy distinguished: monetary policy operates on the money supply, interest rates and credit conditions, conducted by the central bank; fiscal policy operates on taxation, expenditure and borrowing, conducted by the government. Their coordination problem is examinable: fiscal expansion financed by borrowing raises interest rates and can offset monetary easing, and a central bank with an inflation mandate may tighten against a government's expansionary stance. The institutional arrangements to manage this are the separation of debt management from monetary policy and a statutory framework limiting the deficit.

Public borrowing and public debt: government borrowing from domestic and external sources; the distinction between internal debt, which redistributes within the economy, and external debt, which is a claim on future foreign exchange; the concepts of the fiscal deficit as the total borrowing requirement, the revenue deficit as borrowing to meet current consumption, and the primary deficit, which excludes interest payments and therefore shows the current fiscal stance. The sustainability condition in plain terms: debt as a share of output stabilises when the growth rate exceeds the effective interest rate on the debt, given the primary balance — which is why the same debt level is sustainable in a fast-growing economy and not in a stagnant one. The critique of deficits is crowding out of private investment and the burden of interest on future revenue; the case for them is counter-cyclical stabilisation and the financing of assets whose benefits accrue to future periods.

The budget: an estimate of receipts and expenditure for a financial year, requiring legislative authorisation.

Its functions are four and should be named: accountability, since it is the legislature's authorisation of expenditure and the basis on which it is later audited; allocation of resources between competing claims; stabilisation of the economy; and management of the executive's own operations.

Types and forms, each with its claim and its failure.

Line-item budgeting classifies expenditure by object — salaries, travel, supplies. Advantage: simple, and excellent for control of inputs. Failure: it says nothing about what the money achieves.

Performance budgeting classifies by function, programme and activity, linking expenditure to physical outputs and unit costs. Advantage: connects money to work done. Failure: it measures outputs rather than outcomes, and outputs are gamed where they are the basis of allocation.

Programme budgeting and planning-programming-budgeting organise expenditure by objective across departmental boundaries, with multi-year costing and explicit alternatives. Advantage: it makes trade-offs visible. Failure: heavy analytical requirements, and it cuts across the departmental structure through which money is actually appropriated and controlled, which is why it was abandoned in most places that tried it.

Zero-base budgeting requires every activity to be justified afresh each cycle rather than incrementally from the base, through decision packages ranked by priority. Advantage: it exposes activities surviving only by inertia. Failure: the analytical burden is very large, most expenditure is committed by law or contract and cannot in fact be zero-based, and it is therefore rarely sustained beyond a few cycles.

Outcome budgeting attaches outcome indicators to allocations. Advantage: it directs attention past outputs. Failure: attribution of an outcome to an allocation is difficult, indicators are chosen for availability rather than validity, and the framework becomes a reporting exercise where the underlying data is weak.

Gender and other targeted budgeting statements report the share of expenditure benefiting a defined group, which makes distributional effects visible without altering the appropriation structure.

The budgetary process, given as a cycle with the point of control at each stage.

Formulation: estimates prepared by spending units, scrutinised by the finance ministry against resource availability, and settled through negotiation. The control point: the finance ministry's scrutiny. The weakness: incremental anchoring on the previous year's base.

Legislative authorisation: presentation, general discussion, detailed consideration of demands for grants, voting, and the appropriation and finance legislation. The control point: the legislature's power to refuse or reduce. The weakness: compressed scrutiny — most demands are not discussed individually and are passed by the guillotine when time expires, so the formal power of refusal is rarely exercised in practice.

Execution: sanction, allocation, expenditure against rules, re-appropriation between heads within limits, and supplementary provision where the original is insufficient. The control point: the sanction and the internal audit. The weaknesses: unspent allocations, reflecting weak absorption capacity; the end-of-year rush, in which unspent money is spent hurriedly to avoid lapse, with predictable effects on quality; and re-appropriation used to defeat the legislature's allocation.

Accounting and audit: recording of transactions, preparation of accounts, and independent audit reported to the legislature. The control point: the audit report and its examination by the public accounts committee. The weaknesses: audit is after the fact, so it identifies loss rather than preventing it; the volume of transactions means only a sample is examined; and the interval between the transaction and the committee's report weakens the consequence.

Financial accountability is therefore best presented as a chain: legislative authorisation before expenditure, executive control during it, and audit and legislative examination after it — with each link having a characteristic weakness, and with the overall system's strength determined by the weakest.

Types of audit worth naming: regularity or compliance audit, checking that expenditure conformed to authorisation and rules; propriety audit, examining whether expenditure was wise and appropriate even if formally regular; and performance audit, examining economy, efficiency and effectiveness. The objection to performance audit is that it moves the auditor toward evaluating policy, which is the executive's and the legislature's function; the reply is that it examines the execution of policy rather than its choice, and the boundary is genuinely contested.

9. Worked answer — a 20-mark question

Question: "Legislative control of administration is formally comprehensive and practically weak." Examine with reference to financial control. (20 marks)

Framing. The proposition is largely correct and requires qualification in one direction: the formal architecture of legislative financial control is complete, covering the entire cycle from authorisation to audit; its practical weakness is uneven across the stages, being severe at authorisation and considerably less so at post-audit examination. The explanation is structural rather than a matter of diligence.

The formal architecture. No money may be drawn from the consolidated fund without appropriation by law; the annual budget must be laid before the legislature; demands for grants are voted individually and may be refused or reduced; taxation requires legislative authority; the audit of all government expenditure is conducted by an authority independent of the executive and reported to the legislature; and the financial committees examine the audit reports, the estimates and the accounts of public undertakings. Every stage of the expenditure cycle has a legislative check attached to it.

Where the practice is weak, and why.

At authorisation, severely. The time available for discussion of demands for grants is short relative to their number, so most demands are not discussed individually, and the remainder are passed by the guillotine when the allotted time expires. The formal power to refuse or reduce a grant therefore exists and is almost never exercised, because a reduction motion is treated as a challenge to the government and party discipline defeats it. The legislature's real function at this stage is debate on policy, not the scrutiny of allocation.

During execution, structurally. Control passes to the executive. Re-appropriation between heads within a grant is an executive power that alters the distribution the legislature approved. Supplementary demands authorise expenditure after the need has arisen, and their retrospective character makes refusal impractical. Unspent balances and the end-of-year rush are visible only afterwards.

At audit, least weak — and this is the qualification the proposition needs. Audit by an authority independent of the executive, reporting to the legislature, is the one point where the executive does not control the process, and the public accounts committee's examination of audit findings, conducted with a convention of non-partisan working and a chair from the opposition, has produced substantive accountability outcomes. Its limits are real but different in kind: audit is after the fact, so it establishes loss rather than preventing it; only a sample of transactions is examined; the interval between transaction and report is long; the committee's recommendations are not binding; and the number of reports far exceeds the committee's capacity to examine them.

The structural explanation, which is the answer's core. Legislative financial control is weak at the stages where the executive commands a majority in the body exercising the control, and stronger where the process is conducted by an independent authority or by a committee operating outside the majority-minority division. Authorisation is a plenary function subject to party discipline; audit is conducted by a constitutionally protected authority; and the public accounts committee works by convention as a non-partisan body chaired by the opposition. The pattern is exactly the one visible across administrative reform generally: mechanisms vesting authority in an independent office work, and mechanisms depending on the political majority to restrain itself do not.

What follows for reform. Proposals to strengthen the plenary stage — more days for demands, restrictions on the guillotine — attack the symptom, because the constraint is not time but incentive. Proposals with better prospects operate on the independent points: strengthening committee capacity with research support, pre-budget scrutiny by departmental committees, which examine demands outside the plenary and outside the guillotine, reducing the audit-to-report interval, and requiring action-taken reports with consequences attached. Each of these adds capacity to a point where the majority's incentive does not determine the outcome.

Assessment. The proposition stands, with the qualification that "practically weak" understates the audit and committee stage. The accurate statement is that legislative financial control is weakest before the money is spent and strongest after, which is the reverse of what would be desirable, since prevention is worth more than detection — and that inversion, rather than any deficiency of formal power, is the real problem.

Common traps

  • Treating accountability and control as synonyms; they differ in direction and in time.
  • Describing legislative, judicial and executive control separately without the comparison by reach, which is what the question wants.
  • Giving Dicey's position without the critique of it, or presenting administrative law as opposed to the rule of law.
  • Naming Riggs's three concepts without applying them to a specific Indian divergence.
  • Treating the anti-development thesis as a position to be endorsed rather than as a critique to be assessed.
  • Presenting e-governance benefits without the digital divide and exclusion-error costs.
  • Listing budget types without the failure mode of each, which is where the marks are.
  • Treating the delay in disciplinary proceedings as an inefficiency rather than as the price of the protection that secures independence.

Memory aids

  • Finer against Friedrich: external control against internal responsibility; both right about the other's limits.
  • Control by reach: legislature after the fact, judiciary on legality, executive internally, citizen at delivery.
  • Riggs's three: formalism is prescribed against practised; heterogeneity is coexistence; overlapping is the social penetrating the formal.
  • The four budget failures: line-item ignores results; performance measures outputs; PPBS cuts across appropriation; ZBB cannot zero-base committed expenditure.
  • The budget cycle's control points: finance ministry at formulation, legislature at authorisation, sanction at execution, audit after.
  • RTI's mechanism: an individually claimable right with a penalty on the withholder — which is why it worked.

Exam protocol

  1. On accountability questions, compare mechanisms by what each can reach rather than describing them serially.
  2. On administrative law, give Dicey and the critique of Dicey.
  3. On comparative administration, apply the prismatic concepts to a named Indian divergence.
  4. On development, trace the movement of the concept rather than describing the activity.
  5. On policy, distinguish the stage at which the question's failure occurs before prescribing.
  6. On budgeting, give each form with its failure mode, and locate the control point in the cycle.
  7. Close by identifying where the mechanism's independence lies, since that is what predicts whether it works.

Key formulas & results

Everything to memorise for the exam hall, in one card. Screenshot this for revision.

Control by reach
The comparison, not the description, is what accountability questions are asking for.
Riggs's prismatic features
Prescribed against practised; coexistence of unlike systems; the social penetrating the formal.
Debt sustainability
Why the same debt level is sustainable in a fast-growing economy and not in a stagnant one.
The budget cycle's control points
Finance ministry, legislature, sanction, and independent audit — each with a characteristic weakness.
⚠️

Traps UPSC CSE sets — and how to dodge them

These are the exact option-traps and misreads that cost marks under negative marking.

WATCH OUT
Using accountability and control as synonyms.
Accountability is the retrospective obligation to answer to someone with a right to require it; control is the prospective capacity to direct or restrain. An official can be controlled without being accountable and the reverse.
WATCH OUT
Describing legislative, judicial and executive control serially.
Compare them by reach. Each catches what the others cannot, none reaches the point of delivery, and that gap is why the citizen instruments exist.
WATCH OUT
Giving Dicey's position as though it settled the matter.
He misdescribed droit administratif, ignored discretion already conferred by English statute, and could not accommodate the welfare state. Administrative law is the rule of law applied to administrative power, not its negation.
WATCH OUT
Naming Riggs's three concepts without applying them.
Formalism explains the gap between statutory entitlement and delivery; heterogeneity explains why one statute produces different administration in different districts; overlapping explains why formal posting rules coexist with informal influence.
WATCH OUT
Presenting e-governance as an unqualified gain.
Computerising without redesigning encodes the existing complexity; the digital divide excludes those who most need the service; and authentication exclusion errors deny entitlements to people who hold them.
WATCH OUT
Listing budget types by their features.
Each is defined by what it fails to do. Line-item ignores results, performance budgeting measures outputs rather than outcomes, PPBS cuts across the appropriation structure, and zero-base cannot in fact zero-base committed expenditure.

Exam-pattern practice

PYQ-style questions with full solutions. Work through them as a readiness check — mark yourself honestly and get your gap report at the end.

Readiness check

Are you exam-ready for Accountability, Law, Policy & Financial Administration?

15 problems from this chapter. Try each one, reveal the worked solution, mark yourself honestly — get your gap report at the end.

15 questions~11 min

5-minute revision

The whole chapter, distilled. Read this the night before the exam.

  • Accountability is retrospective answering; control is prospective direction; responsibility is internal
  • Finer external against Friedrich internal — each right about the other's limits
  • Control by reach: legislature after the fact, judiciary on legality, executive internally, citizen at delivery
  • Judicial review's three grounds: illegality, irrationality, procedural impropriety
  • RTI worked because it creates an individually claimable right with a penalty on the withholder
  • Dicey: no arbitrary power, equality before the law, constitution as result of ordinary law — and the three critiques
  • Delegated legislation controlled legislatively by laying, judicially by ultra vires, procedurally by publication and consultation
  • Essential legislative function: policy and standard must be laid down by the legislature
  • Riggs: fused, prismatic, diffracted; formalism, heterogeneity, overlapping; sala between chamber and bureau
  • CPA critique: descriptive not predictive, explains after the fact, Western endpoint, static
  • Development concept: growth, growth with distribution, basic needs, capabilities, sustainability
  • Bureaucracy and development: Weberian features raise capacity and reduce responsiveness — hence the two-form landscape
  • SHG mechanism: group liability substituting for collateral; organisational effect exceeds the financial one
  • Implementation failure: clearance chains, assumption gaps, street-level coping, resource mismatch, goal ambiguity, capture
  • Budget failure modes: line-item ignores results, performance measures outputs, PPBS cuts across appropriation, ZBB cannot zero-base commitments
  • Financial control is weakest before spending and strongest after — the inversion is the real problem

UPSC CSE question blueprint

How this topic is asked, tier by tier — so you can prep to the pattern.

Typical weightage: 250

Question styleMarks eachTypical countWhat it tests
Accountability, control and administrative lawReliably examined every year, usually comparatively
Comparative and development administrationRiggs and the development debates are standing questions
Personnel administration and public policyFrequently set, with implementation the most examined stage
Improvement techniques and financial administrationE-governance and budget forms are the highest-yield items
Prep strategy
  • Hold the control-by-reach comparison as a single prepared page
  • Prepare Dicey with his critique, and Riggs with the application and the critique
  • Learn each budget form by its failure mode rather than its definition
  • Prepare the implementation failure list as a diagnostic, since it answers many questions
  • Keep development content to the movements and debates, not the programmes

Exam-hall strategy

Battle-tested tips from mentors and toppers for this topic under the sectional clock.

  1. Compare accountability mechanisms by reach rather than describing them serially.
  2. Give Dicey and the critique of Dicey together; never present administrative law as opposed to the rule of law.
  3. Apply Riggs's three concepts to a named Indian divergence rather than defining them.
  4. State the anti-development thesis as a critique to be assessed, not a position to endorse.
  5. Hold both findings on bureaucracy and development and explain the two-form landscape they produce.
  6. Locate a policy failure at its stage before prescribing a remedy for it.
  7. Give each budget form with its failure mode and each cycle stage with its control point.
  8. Close by identifying where the mechanism's independence lies, since that predicts whether it works.

Beyond the exam

Where this skill shows up in the job you're competing for — and in life.

Choosing an accountability instrument

Match the mechanism to the class of failure — legality to review, delivery failure to a claimable right — rather than adding oversight generally.

Reading why a scheme underperforms

Clearance count, assumption gaps, street-level coping and capture are checkable causes with different remedies.

Assessing a computerisation proposal

Ask whether the process is being redesigned or encoded, and what happens to those the digital channel excludes.

Where else this topic is tested

Prepare once, score in every exam that asks it.

UPSC CSE Public Administration Paper IIFinancial management, administrative reform, civil services and accountability institutions apply this material directly
UPSC CSE Mains GS Paper IITransparency, accountability institutions, statutory bodies, policy design and welfare delivery overlap substantially
UPSC CSE Mains GS Paper IIIBudgeting, fiscal policy, public expenditure and government intervention draw on the same content

Questions aspirants ask

Pulled from the Q&A community and mentor sessions.

Organise by reach rather than by institution. Open with the proposition that each mechanism catches a different class of failure and none catches all of them, then take the classes in turn: legality and procedure are reached by the judiciary; gross financial failure by audit and the legislature; systemic and routine failure by internal executive control, which is also where that control is weakest because it is the administration checking itself; and failure at the point of delivery by the citizen instruments, which exist precisely because nothing else reaches there. Close with the structural finding that mechanisms working independently of the political majority's attention — independent audit, judicial review, individually claimable rights — perform, and those depending on the majority to restrain itself do not. That structure answers questions on legislative control, on judicial control, on citizen-administration and on accountability generally, with the emphasis adjusted, and it is an argument rather than an inventory.

The control architecture and its failure points, not the accounting detail. What earns marks: the budget's four functions; each budget form with the specific thing it fails to do; the four stages of the cycle with the control point and weakness of each; the difference between regularity, propriety and performance audit with the objection to the third; and the fiscal, revenue and primary deficit distinction with what each tells you. What earns nothing: the classification codes, the detailed heads of account, the precise stages of the finance bill, and the current year's numbers, which date immediately and are not what a theory paper is testing. The single most useful framing to carry in is that financial control is a chain — authorisation before, sanction during, audit after — whose strength is that of its weakest link, and that the Indian chain is inverted, being weakest before the money is spent.

Less than they appear, because the units are examined for movements and debates rather than for the substance of development itself. Four things carry almost all the marks. First, the movement of the development concept from growth through distribution and basic needs to capabilities and sustainability, with the reason for each broadening. Second, the changing profile of development administration from state-as-agent and capacity-as-constraint to decentralisation, participation and rights-based entitlement, with the evidence that forced each shift. Third, the two findings on bureaucracy and development held together, and the two-form administrative landscape they explain. Fourth, the strong state versus market debate resolved as a question about state capacity rather than about state or market. The anti-development thesis and women and development are each worth a compact prepared treatment. What is not examinable is the substance of any particular development programme, which belongs to Paper II and to General Studies.

Header Logo