AN-008: Placebo on never-litigated items

Economic intuition

If the price gap were really driven by something generic, a platform-wide pricing quirk or a common time trend, it should also appear on items that are never litigated. It does not. The placebo coefficient on never-litigated items is essentially zero, so the regime contrast is not a mechanical feature of the platform or the calendar.

Question

A concern with the regime contrast is that it could reflect a platform-wide pricing pattern or a common time trend rather than anything specific to litigation. If so, a comparable gap should surface on items that are never litigated. This page tests for that spurious gap.

Design

• Sample: never-litigated items within the BEC group 65 São Paulo pharmaceutical sample, 2009–2019.
• Specification: a negotiated-price placebo regression estimated on items that are never subject to litigation, testing whether a spurious gap arises where none should.
• Role: a falsification check against generic platform or time-trend explanations; it is not a stand-alone identification design.

Results

Quantity	Coefficient	SE
Negotiated-price placebo	−0.020	0.032

Output: v10-causal-mechanism/output/tables/tab_placebo.tex.

Interpretation

Confidence: yellow. The negotiated-price placebo coefficient is −0.020 with a standard error of 0.032: economically and statistically null. A generic platform-wide or time-trend mechanism would have produced a non-trivial gap on never-litigated items; the absence of one falsifies those explanations. This is a falsification check, not a stand-alone identification design. The reading is yellow because the evidence is from a single jurisdiction (São Paulo BEC) and rests on own project runs.

Follow-ups

• Pair this falsification with the dynamic event-study and Honest-DiD sensitivity diagnostics in AN-010, which together populate the placebo-and-dynamics hypothesis.
• Re-estimate the placebo within narrower therapeutic subgroups to confirm the null is not masking offsetting subgroup effects.