Problem diagnosis and decision boundaries · Lesson 1

From the visible symptom to the regulatory problem

5 min read

Lesson objectives

  • Separate a service delay from its possible causes, test competing explanations, and write a bounded problem statement before selecting an instrument.
Reading tools

A policy team receives a report that activity permits are delayed. Delay may merit action, but it neither explains its own cause nor automatically justifies a new rule. First define the unit of measurement. Does the clock start with a complete application, or with a first visit to the portal and end at issuance? Mixing incomplete and complete applications changes the comparison. Next define the affected population: all applicants or small firms in one district? Without these definitions an overall average may improve while the most affected group fares worse. In regulatory policy analysis, frame the problem as a testable gap between current conditions and a public outcome, then investigate causes before choosing an instrument.

Draw a short causal chain rather than moving from complaint to remedy. Inputs might include staff capacity, clarity of requirements and application quality. The process includes screening, clarification, referral and decision. Processing time is an immediate output; the wider outcome may be entry into a market without weakening consumer protection. Ask which link best explains the observed delay. A high share of incomplete applications points towards guidance or form design. A backlog of complete files after referral points towards a coordination bottleneck. Divergent decisions on similar files may signal an ambiguous criterion or inconsistent application. These are hypotheses, not legal findings. Examine sample files and stage dates before attributing delay to statutory wording.

Distinguish three kinds of evidence. Transaction records show what happened inside the system but can miss people who abandoned an application before submitting it. Applicant interviews reveal comprehension costs and repeated steps, yet recall may be inaccurate and vocal applicants unrepresentative. A before-and-after comparison shows change, not causation, if volume, staffing or recording methods also changed. Record conflicting evidence in a table: observation, source, coverage limit, alternative explanation and next test. A sound early recommendation might be to trial clearer guidance and collect data on the critical stage before proposing a new obligation. If the legal text itself mandates an unnecessary step, assessment of an amendment belongs with the authorized body.

A problem statement separates observation, goal and assumption: 'In the defined sample, complete applications wait longer after referral between two bodies; we do not yet know how much the document rule contributes compared with processing capacity; we seek less delay while preserving risk screening.' This allows several options: simplify a form, clarify guidance, improve administrative links, consider a legal amendment after formal verification, or retain the status quo if change costs exceed benefits. Include a discriminating test. If most waiting precedes completeness, adding decision staff alone will not solve it. If delay is concentrated in a legally required security check, removing the check simply to improve the average would be unsound; both authority and impact require scrutiny.

Lesson summary

Inputs: 40 complete applications at six days and 20 returned applications with eight days before completeness plus six after. Solution: the complete group takes six days; the returned group takes 8+6=14 days. If an overall average is needed, it is (4