Duplicative Funding and Funding Gaps

1. 1. Federal Consolidated Revenue Fund → AB Province (major transfers)

   Partial

   What we see

   Finance Canada major-transfer tables (CSV, monthly cash flow) tell us how much CHT, CST, ICIP, CCBF, etc. Ottawa sent to Alberta each year back to 1980.

   What we don’t

   The amount on the Finance Canada side does not reconcile to a matching line on the Alberta budget side. AB's fiscal-plan revenue category 'Federal transfers' is one aggregate figure; the per-program decomposition lives at the federal end only.

   Example question

   Answerable: 'How much CHT did Ottawa send AB in FY 2026–27?' (yes — Finance Canada CSV.) Unanswerable: 'Where in the AB fiscal plan does that $7B appear?' (only as part of an aggregate revenue line.)

2. 2. AB Province → AB Ministries (estimates + fiscal plan)

   Visible

   What we see

   budget_estimate_lines and budget_fiscal_plan_lines give per-ministry voted-expense and capital allocation by FY back to 2000 (extended to 25 years by spec 017's PDF parser).

   What we don’t

   Federal transfers stay aggregated on the revenue side; there is no 'this $X of Health came from CHT' attribution.

   Example question

   Answerable: 'What was Health's voted expense in FY 2026–27?' (yes.) Answerable: 'Has it grown faster than total program expense over 10 years?' (yes.)

3. 3. AB Ministries → Disclosed payments (reconciliation gap)

   Partial

   What we see

   ministry_budget_vs_disclosure computes a gap_pct per ministry × FY: what was budgeted vs what shows up in disclosure streams. Spec 016 + 043 closed the ministry-name aliasing problem (data-quality issue DQ-001) so the comparison is now meaningful.

   What we don’t

   Sub-threshold payments (under disclosure cutoffs) and intra-government transfers don't appear in disclosure, so the gap is structural, not just a data error. Crown corp opex flows out of ministries but is reported separately.

   Example question

   Answerable: 'What % of Advanced Education's FY 2025–26 budget reconciles to published disclosures?' (yes — gap_pct.) Unanswerable: 'What's in the unreconciled portion?' (only inferable from line aggregation.)

4. 4. AB → Recipients (grants, contracts, payments)

   Visible

   What we see

   grant_payments, blue_book selected payments, sole_source_contracts, travel/p-card disclosure. Recipient-resolved via the deterministic entity-resolution layer (spec 043)'s registry-first entity resolver. ~5,254 minister-office rows now person-attributed via spec 016.

   What we don’t

   Recipient names are not stable IDs. 'University of Calgary' and 'U of C' are the same entity only after resolution; the upstream publisher does not provide a key.

   Example question

   Answerable: 'Top 50 vendors to AB Health by FY 2024–25 dollar volume?' (yes.) Answerable: 'Has any vendor's share of sole-source spend grown >5pp YoY?' (yes — joins blue_book × sole_source_contracts.)

5. 5. Recipients → Sub-recipients (first hop only)

   Partial

   What we see

   AHS publishes a vendor disclosure (PDF). School board audited financial statements list payees above thresholds. Municipal SOFI reports salaries + payments above $50k. ab_spending has not yet ingested any of these (specs 027, 028, 029 in plan).

   What we don’t

   When the AB government grants $X to AHS, where does AHS spend it? When AB grants $Y to a school authority, who does the authority pay? The first-hop disclosure is partial; format is per-entity PDF, not bulk CSV.

   Example question

   Partially answerable after spec 027 ships/028: 'Which AHS vendors received the largest share of FY 2024–25 grant inflow?' (PDF parse required.) Unanswerable today: 'What % of municipal SOFI spend can be linked back to provincial transfers?' (no joinable entity ID across layers.)

6. 6. Sub-recipients → Sub-sub-recipients

   Invisible

   What we see

   Nothing systematic. A school board's audited financials show vendor payments but the vendor's onward spending is private. A municipality's SOFI shows contractor payments but the contractor's sub-contractors are invisible.

   What we don’t

   The legal disclosure regime ends at the first level of public spending; private-sector sub-flows are out of scope for any FOIP / public-finance statute in Canada.

   Example question

   Unanswerable: 'How much of the federal CCBF infrastructure dollar that became a Calgary project actually flowed to Alberta-based subcontractors vs. out-of-province ones?' (no public dataset spans the contractor layer.)

7. 7. End-use outcomes (did the money do what it said it would?)

   Invisible

   What we see

   CIHI publishes aggregate health-system outcomes. ESDC publishes labour-market participant outcomes. Per-program outcome data is occasionally published in evaluations.

   What we don’t

   The dollar-to-outcome chain. Even where outcomes exist (hospital wait times, employment-12-months-post), they cannot be reconciled with the dollars that funded them — because the dollars are not traceable past layer 5.

   Example question

   Unanswerable in any rigorous sense: 'Did the $7B of CHT inflow improve AB health outcomes vs. what would have happened without it?' Aggregates exist but the counterfactual link does not.

• Federal→province layer (not ingested; only aggregate on fiscal plan revenue).
• Province→sub-recipient cascades (AHS, school boards, municipalities stop at first hop).
• Sub-entity drill-down for Crown corps (data-quality issue DQ-003).
• Decomposition of AB fiscal-plan aggregate revenue into per-program federal sources.
• Attribution rules for conditional grants (CHT, ICIP, CMHC bilaterals).

See chapter 11 · the data-gaps matrix for the full inventory.