a study about information and authority

Who sees, who decides?

In 2015, China moved decision rights for some branches from the centre to prefecture supervisors. Follow the same information through report, centre and intervention.

documented bank case · model mechanism · transferable question

enter the file

the stake

The change was not only geographic

The 2015 reform did not create autonomous regional regulators. It moved decision rights over local-bank branches to prefecture-level supervisors while keeping them accountable to the centre. That separates the question about information from a slogan about decentralization.

The reader sees what the study reports first. Only then comes a reconstruction of the proposed mechanism and a transfer sheet for other systems.

result in the data
mechanism in the model
measurement limit

the central object

Move the stamp, keep the hierarchy

Choose who decides in the same scene. The observation stays local. The point where information becomes actionable changes.

Choose the decision authority
model reconstruction

Authority is at the prefecture. All three file fragments are available at the point that can order the intervention. Accountability to the centre remains.

The two architectures without JavaScript

The scene explains the mechanism. The study reports aggregate results, not public before-and-after matches for individual cases.

The two architectures without JavaScript
decisionOld architectureDelegated decision
local observationprefectureprefecture
decisioncentral authorityprefecture authority
Accountabilitycentrecentre
What happens to detailRequires transmission and interpretationAvailable at the decision point

what the authors find

More interventions. A more differentiated pattern.

If delegation only meant a tougher local hand or suddenly riskier branches, we should see a different profile. The authors look for those alternatives.

  1. 01observed result

    More interventions

    Branches affected by the reform were 57–80% more likely to receive a supervisory intervention.

    The detailed estimate is +6.3 to +8.8 percentage points, against an unconditional probability of 11.1%.

  2. 02observed result

    Less uniform fines

    Average fines fall while fine dispersion rises. The pattern is consistent with detecting less severe cases and making more differentiated responses.

    This is a result pattern, not proof that every additional intervention was a real violation newly discovered.

  3. 03observed result

    Information distance matters

    Effects are larger where the centre’s informational disadvantage is more plausible, for example farther from Beijing.

    These heterogeneous effects fit the authors’ proposed mechanism, but do not directly observe what was lost in an individual report.

  4. 04downstream behaviour

    Lending changes

    Later, local branches charge loan spreads about 31 basis points higher and make smaller loans.

    This behaviour is consistent with more conservative risk-taking. It does not show that the result is socially optimal.

  5. 05unknown

    What we cannot see

    The data do not contain the universe of real violations that went undetected or unsanctioned.

    The improved-detection interpretation and any efficiency gain remain limited by the model and by what the design can measure.

How to read the first number11,1%unconditional intervention probability
estimated increase after reform+6,3–8,8 pppercentage points, not percentages

The difference between a recorded intervention and a real violation matters. The dataset sees enforcement, not the universe of events that never reached the register.

the question that travels

Transfer the question, not the result

Choose a domain. The sheet does not say who should decide and does not assign China’s effect to Romania. It shows what to verify before proposing a move in authority.

Choose a domain to test

selected domainbank supervision

The starting case: the bank, branch, prefecture and centre hold different information and rights.

evidence to look for

intervention register, decision authority, spreads and loan size

working formulation

In this domain, the next comparison worth checking is: information appears at the branch, decision rights are prefectural, and accountability remains central. The next evidence needed is the exact file path and what reaches the intervention decision. This is not a verdict about the institution.

  • environmental inspection

    The evidence may be local while the cost of pollution crosses a jurisdictional boundary.

    inspection register, sanctioning authority, appeals path
  • public health

    The clinical signal appears close to the patient, while consistency and accountability may be centralized.

    protocol, escalation path, feedback after a decision
  • tax administration

    Data may be fine-grained locally while rules and audit are uniform at the centre.

    what can open locally, what the centre approves, what stays in the file
  • building control

    An inspector sees the site, while permits, sanctions and neighbours may require coordination.

    stage-level authority, inspection record, effects on neighbours
  • social services

    Case knowledge is personal, while decisions may depend on shared criteria and budgets.

    who defines need, who approves support, who checks the result
  • public procurement

    Market context is local, while comparison and oversight may be designed nationally.

    market data, thresholds, review and complaints path
  • bank supervision

    The starting case: the bank, branch, prefecture and centre hold different information and rights.

    intervention register, decision authority, spreads and loan size
five questions for any hierarchy

There is no score. Local authority is not automatically better, and centralization is not automatically worse.

  1. 01

    Where does the highest-resolution observation appear?

    Find where the case can be seen before it becomes a category.

  2. 02

    Who can turn an observation into a case?

    Separate the right to report from the right to open a procedure.

  3. 03

    Who can decide the intervention?

    Name the formal authority, not only the institution on the chart.

  4. 04

    What must travel upward before action can begin?

    Note what becomes a summary, category or threshold.

  5. 05

    What preserves control and accountability?

    Check feedback, audit, capture and effects beyond the territory.

The question stays open: how much of what the local office sees can be used before it becomes only a summary?

sources and limits

Three levels of evidence

The page separates the empirical result, the author’s public summary and the open draft used for detailed checks. Transfer to other domains is a research proposal, not a finding of the study.

  1. S1 · primary source

    CEPR Discussion Paper 21846

    Published 16 August 2026. A discussion paper, not peer reviewed. Supports the reform, sample, aggregate results and main limits.

    open source
  2. S2 · author page

    Wouter Dessein · publications

    The author’s public summary describes the model of bias and information loss and the distinction between delegation within a hierarchy and autonomous decentralization.

    open source
  3. S3 · open draft

    Authors’ November 2025 draft

    The open version used to check the detailed estimate, empirical architecture and tables referenced by the CEPR summary.

    open source
What this page cannot say
  • It does not show the universe of real violations that went undetected.
  • It does not show that local authorities were less biased.
  • It does not show that more interventions mean a safer society or socially optimal supervision.
  • It does not estimate what would happen in Romania or another domain.
  • It does not turn a model reconstruction into a record of an actual report.
reading method

Facts about the reform and aggregate results are marked as documented. The hypothesis about information being translated or lost is marked as a model mechanism. Questions for other systems are marked as transfer.