Findings — Sourcing under Sanctions¶

This page is the curated index of what we have learned from the project's own analyses. Each finding is a claim about the world — the kind of statement that would go in the paper — and it may rest on several AN-NNN analyses, cited in its Sources footer. The detail, design, and caveats behind each number live on those analysis pages; this page is the synthesis.

This page is the directory of conclusions. Use docs/analyses/ to look up the per-result design, raw numbers, and full result tables; use this page to scan what we believe and at what confidence.

How to read the confidence tags¶

A traffic-light convention runs across the findings: 🟢 = strongest confidence, 🟡 = middle, 🔴 = weakest.

Empirical findings — the color reflects the source of confidence, not the size of the effect:

🟢 Replicated — the finding appears in multiple independent samples or jurisdictions that agree in direction and rough magnitude.
🟡 Single source — one solid study, or one of our own runs, with no independent replication yet.
🔴 Provisional — one descriptive cut that is sample-definition-dependent or flagged with a known caveat.

Interpretations — parallel scheme: 🟢 strong (converging lines, alternatives considered and rejected), 🟡 plausible (consistent but other readings open), 🔴 speculative.

Every finding here is currently 🟡: each rests on the project's own runs of the São Paulo BEC pharmaceutical sample (2009–2019), none has independent replication in another jurisdiction, and the central contrast disciplines — but does not eliminate — selection into the administrative urgent channel, which is a selected, larger, closest-feasible comparison rather than a randomized one.

Findings overview¶

The urgent-procurement environment (empirical, motivational)

Legal urgency raises procurement costs and weakens competition — urgent purchases carry ~5.4% higher negotiated prices, ~5.4% fewer bidders, and a ~2.1pp higher tender-success rate than ordinary purchases of comparable items.

The sanction-related margin: pricing versus sourcing (empirical + interpretive)

The under-the-gun gap is selection-bounded at 15.9–21.1% — court-mandated urgent purchases are costlier than the closest administrative urgent comparison; after Lee trimming for selection, the litigated-over- administrative gap is bounded in [15.9%, 21.1%].
No broad same-firm markup in deep repeated urgent markets — within the same firm, buyer, and item, prices are statistically indistinguishable across urgent regimes (0.035, SE 0.041). A deep-market null, not a universal claim.
The deep-market null is not universal — a residual within-firm gap persists in the earlier period (+0.117, robust to a quantity control); the quantity dimension reflects scale, not same-firm pricing.

The sourcing mechanism (empirical + interpretive)

Fragmented sourcing is the margin — administrative urgent orders are ~3.3× larger (lost scale), and the modal winning supplier differs in 70.2% of item-buyer pairs (supplier-set reallocation).
A judicial-enforcement form of passive waste — one-sided sanctions secure delivery but weaken the routines that aggregate demand and match buyers with suppliers.

Fiscal scope (interpretive)

Fiscal procurement-cost implication of ~$27.8M per year — applying the bounded gap to a calibrated admissible share of litigated spending; a fiscal procurement-cost calculation, not a full welfare estimate.

Open items for this page¶

Nothing is 🟢 yet. Every finding is single-source (our own run on the São Paulo BEC sample). Promotion to 🟢 would require an independent replication in another procurement jurisdiction or a second identifying source within São Paulo.
Selection is bounded, not eliminated. The administrative urgent channel is selected and larger; the Lee bounds discipline that selection under a monotonicity restriction but do not recover what a litigated purchase would have cost without sanctions. See H:utg-gap-selection-bounded.
The deep-market null is not universal. A residual within-firm gap persists in the earlier period (the quantity dimension instead reflects scale); the no-broad-same-firm-markup reading is specific to deep repeated urgent markets. See The deep-market null is not universal.
The fiscal figure is a procurement-cost calculation. It is not a full welfare estimate and excludes patient health benefits, search and compliance costs, and other welfare components.