Policy Learning and Policy Failure: Definitions, Dimensions and Intersections
Bibliographic record
Abstract
Introduction The social security system in the UK has long been regarded as overly bureaucratic and too complex for either bureaucrats or claimants to entirely understand. In 2010, the Conservative-led coalition government unveiled Universal Credit (UC) as the answer to this historic policy problem. An ambitious plan to merge six in-and outof-work benefits, UC aims to ensure not only a simpler system but also that it is more beneficial to be in work than on the dole. Entailing a huge administrative challenge, the implementation of UC has been dogged by problems from the outset; with costs spiralling and the timetable slipping, the policy was effectively ‘reset’ in 2013 and political pressure to abandon it mounted in the months that followed. Yet, 2013 proved to be a turning point. Looking into the precipice of the failure of a flagship reform, policy makers engaged in policy learning. Along with analysing the technical problems and capacity deficits they faced, civil servants learned from previous experiences of implementing complex policies and from similar problems in social security reform in Australia. Appointing senior troubleshooters responsible for getting the programme on track and engaging a recovery team – Major Projects Authority (MPA) – the UC appeared to have been turned around (see Timmins, 2016 for a full account). But such optimism was premature. Since its staggered roll-out, the Resolution Foundation think tank estimated UC will result in 3.2 million working families being worse off by an average of £48 a week. As many as 600,000 claimants are no longer entitled to any assistance at all (Finch and Gardiner, 2018). In April 2018, the Trussell Trust charity reported an average 52 per cent rise in demand for its food banks in areas where UC had been rolled out (Trussell Trust, 2018). Though reputed by government at the time, in February 2019 the sixth Secretary of State for Work and Pensions presiding over UC – Rt Hon Amber Rudd – conceded that its introduction had contributed to increased food insecurity (HC Deb, 11 February 2019, c593).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.046 | 0.054 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.011 | 0.020 |
| Science and technology studies | 0.015 | 0.165 |
| Scholarly communication | 0.043 | 0.048 |
| Open science | 0.006 | 0.031 |
| Research integrity | 0.016 | 0.023 |
| Insufficient payload (model declined to judge) | 0.010 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".