Paper I

Story hook

In 1966 the United States Economic Development Administration announced a jobs programme for Oakland, California. Around $23 million was committed to build an aircraft maintenance hangar, a marine terminal and an access road, on the theory that public works would create employment for the city's unemployed Black workers. The decision was popular, well-funded, technically feasible and backed by Washington. Three years later, most of the money was still unspent and the jobs created numbered in the dozens rather than the thousands promised. Nobody had stolen anything. Nobody had opposed the policy.

Two Berkeley political scientists, Jeffrey Pressman and Aaron Wildavsky, went looking for the villain and could not find one. What they found instead was a chain of clearances — federal agencies, regional offices, the Port of Oakland, contractors, the General Accounting Office, engineers, lawyers — each of which had to say yes. They counted roughly thirty decision points requiring some seventy separate agreements, and then did arithmetic that has been quoted in every implementation seminar since: if each of seventy clearances has an 80 per cent chance of going through, the probability that all seventy go through is 0.80 raised to the seventieth power — about one in six million. Their own table in chapter 5 of the book prints .000000125 for that row, roughly one in eight million. Even at a heroic 99 per cent per clearance, seventy agreements leave the chance of complete success at about .489 — worse than a coin toss. Their 1973 book, Implementation, carried a subtitle that told the whole story — how great expectations in Washington are dashed in Oakland, and why it is amazing that federal programmes work at all.

India has its own version of that arithmetic, delivered as a single sentence. In 1985, Rajiv Gandhi said that of every rupee the government spent on the poor, only about fifteen paise actually reached the beneficiary. (Quote the year, not a platform: the remark is securely dated to 1985, but the venue is reported inconsistently and its provenance has been contested in print.) The line has been contested, recalculated and repeatedly weaponised ever since — but no Indian administrator has been able to dismiss it. Between that sentence and the Direct Benefit Transfer architecture launched on 1 January 2013, which routes money to a bank account and removes several of Pressman's clearance points by design, lies the whole subject matter of this unit: not what governments intend, but what happens between the intention and the citizen — and how, if at all, we can find out.

That gap is what the UPSC syllabus means when it lists "models of policy-making and their critique" alongside "processes of conceptualisation, planning, implementation, monitoring, evaluation and review and their limitations". The examiner is not asking you to praise a scheme. The examiner is asking you to explain, with named theory, why a good decision so often produces a bad outcome — and what the discipline has proposed to do about it.

Why this matters for UPSC

Public Policy is item 10 of the twelve-item Paper I syllabus of the UPSC Public Administration optional and, in most years, sits in Paper I's Section B; it reliably yields one full question (usually 20 marks) plus at least one 10-mark short note in the compulsory Q1/Q5 blocks, so a realistic budget is 20-40 marks out of 250 in a typical year, and occasionally more when the paper leans towards governance. Questions are almost never factual — they are analytical and comparative ("critically examine", "in the light of", "to what extent"), which is why a candidate who has memorised the six models but cannot say what Dror was arguing against loses marks that look, on the answer script, like they should have been easy. The unit also double-serves: its Indian half (NITI Aayog, DMEO, outcome budgeting, social audit) is directly reusable in GS-II governance answers and in the Essay paper.