## discussion-note-21-boundary-between-government-and-public-nonfinancial-corporations

## Source details

**Canonical URL:** [discussion-note-21-boundary-between-government-and-public-nonfinancial-corporations](https://www.imf.org/-/media/files/data/statistics/gfsm/global-consultations/discussion-note-21-boundary-between-government-and-public-nonfinancial-corporations.pdf)

## Other formats

- [Markdown version](/-/media/files/data/statistics/gfsm/global-consultations/discussion-note-21-boundary-between-government-and-public-nonfinancial-corporations.pdf.md)
- [Structured JSON version](/-/media/files/data/statistics/gfsm/global-consultations/discussion-note-21-boundary-between-government-and-public-nonfinancial-corporations.pdf.json)

---

### Summary and purpose
- The Discussion Note identifies that existing GFSM 2014 guidance on classifying government-controlled nonfinancial units as market or nonmarket producers can lead to uneven implementation and reduced comparability of fiscal data.
- Proposes a more prescriptive framework for the market test combining quantitative and qualitative criteria to strengthen alignment with the 2025 System of National Accounts (2025 SNA).
- Proposed refinements address:
  - Reclassification of compulsory licenses as taxes.
  - The sum-of-costs approach and its expanded components.
  - Guidance on measuring the return to capital.
  - Incorporation of depletion and rent of nonproduced assets as production cost items.
- Task Team Responsible: GFS compilation Task Team (TT1). Authors: Artur Santos, Geraldine Kyalo, Irina Dubinina, Katherine Mills, Bocar Ka.

### Problem statement and 2025 SNA context
- GFSM 2014 paragraphs 2.64–2.75 provide principles for distinguishing market and nonmarket producers but allow differing national interpretations affecting general government coverage and fiscal analysis.
- 2025 SNA changes affecting the market test:
  - Expands costs relevant to the market test to include rent payable on nonproduced assets, depletion, and an explicit net return to capital.
  - Reclassifies compulsory nontransferable licenses as taxes, requiring exclusion from sales revenue used in the market test.
- Eurostat MGDD 2022 and IMF technical assistance introduce more prescriptive methods (decision trees, indicators) combining quantitative and qualitative criteria.

### Issue A — Need for more prescriptive guidance on the market test
Findings
- GFSM 2014 avoids rigid numerical thresholds but notes sales are expected, on average, to be “at least half of the costs of production over a sustained multi-year period,” implicitly suggesting a 50 percent benchmark.
- Most systematic classifications use a market test assessing whether sales cover at least half of production costs over a sustained multi-year period.
- Open methodological questions:
  - Appropriate length of the “sustained multi-year period.”
  - Metric to relate sales-to-cost ratios (average of annual ratios versus consecutive-year rule).
- Sales receipts measurement guidance: measure before product taxes; exclude subsidies on products (unless granted to any producer); exclude payments from government unless paid to similar producers; exclude own-account production.
- Production costs to be captured via the sum-of-costs approach.

Options to strengthen guidance (A1–A4)
- Option A1: Maintain current GFSM 2014 guidance.
  - Preserve principles-based flexibility.
  - Risk: continued cross-country and temporal discrepancies.
  - Note: updates still needed to reflect 2025 SNA sum-of-costs inclusion of rent and depletion.
- Option A2: Decision-tree approach.
  - Adopt MGDD 2022 decision tree; apply qualitative criteria before quantitative test.
  - Use a 50 percent threshold for sales-to-cost ratio calculated over at least three consecutive years.
  - Quantitative test adjustments:
    - Exclude results impacted by one-off or exceptional circumstances.
    - Allow early reclassification if reasonable expectations about future performance exist.
    - Use forecasts of sales-to-cost ratios for start-up entities until sufficient historical data exist.
  - Benefits: greater comparability with preserved flexibility. Risks: over-reliance on single indicator; potential to overlook pricing control, nature of goods/services, influence on market.
- Option A3: Indicator approach (primary and secondary indicators with thresholds).
  - Primary indicators include:
    - Is the government the unit’s sole client? (yes or no)
    - Ratio of sales to production costs over the last three consecutive years.
    - Set of governance/support indicators using 50 percent and 150 percent thresholds (see below).
  - Secondary indicators include profitability, limited competition, and price control (yes/no assessments).
  - Benefits: structured and operationally clear; Risks: reduced flexibility and increased data burden.
- Option A4: Hybrid decision-tree plus indicators.
  - Retain qualitative decision-tree sequencing; apply indicator-based framework (Primary Indicator 2, Primary Indicators 3, Secondary Indicators) before standard sales-to-cost test.
  - Benefits: combines qualitative sequencing with detailed quantitative assessment; Drawbacks: greater complexity and data requirements.
- Draft Task Team recommendation: consider Options A2, A3, or A4 as alternatives to current guidance; all three provide more prescriptive guidance and clearer, structured frameworks for the market test.
- GFS Advisory Committee (GFSAC) October 2025: substantive support for Options A2 and A3; Option A4 developed to address divergent views.

### Practical implementation considerations (Issue A)
- Prescriptive framework must define:
  - Exact components of sales revenue and production costs to be included (e.g., exclude compulsory license revenues reclassified as taxes).
  - Multi-year analysis window (Discussion Note cites three consecutive years in proposals).
  - Procedures to adjust for exceptional one-off events and use of forecasts for start-ups.
- Data and compilation burden: Options A3 and A4 increase data requirements (multiple ratios, three-year averages, debt/equity assessments).
- Risk of misclassification if strict thresholds are applied without contextual assessment (market structure, pricing controls, essential service nature).

### Issue B — Sum-of-costs approach: background and components
- 2025 SNA uses the sum-of-costs approach to value production for own final use and nonmarket production when output is free or prices are not economically significant.
- 2008 SNA components: intermediate consumption, compensation of employees, other taxes less subsidies on production, consumption of fixed capital, and return to capital employed.
- 2025 SNA expands components to include:
  - Depletion.
  - Rent payable for use of nonproduced nonfinancial assets.
  - Adjustments to cost components.
- Production cost components analyzed in the market test include:
  - Remuneration of employees.
  - Use of goods and services.
  - Rent of nonproduced assets.
  - Depreciation.
  - When applicable, depletion.
  - Other taxes on production (without deduction of subsidies).
  - A return to capital employed in production.
- GFSM Research Project 1.27 examined application of sum-of-costs and recommended exploring implications for the market test.

Rent, Depletion, Depreciation
- Rent: income received by owner of a nonproduced asset for making it available to another institutional unit for use in production; 2025 SNA broadens rent to include land rentals, rentals of mineral and energy resources, and use of nonproduced intangible assets (contracts, licenses, goodwill, marketing assets).
- Depletion: reduction in value of nonproduced natural assets used in production; inclusion in 2025 SNA raises costs and may make it harder for extractive-sector entities to meet quantitative market test.
- Depreciation: 2025 SNA renames consumption of fixed capital; distinguish statistical depreciation (current exchange values) from accounting (historical-cost) depreciation; compilers generally rely on financial statements.

Return to capital: conceptual approaches and issues
- 2025 SNA defines net return to capital as rate of return multiplied by the value of nonfinancial assets used in production — shifting from financial-statement expense to applying a rate to asset values.
- Tension: applying a nonmarket producer rate in market test may be inconsistent because market test aims to reflect market revenues and costs.
- WACC conceptually robust but requires inputs often unavailable for nonpublic entities.
- Discussion Note proposes a practical simplified method to preserve conceptual clarity without advanced calculations.

Options for return-to-capital estimation (B1–B3)
- All options incorporate GFSM Update Notice of Decision 1.27: include depletion and rent, and add a GFSM footnote noting financial-statement depreciation/depletion may differ from statistical measures; permit use of financial-statement values for market test practicality.
- Option B1: Net interest as return to capital.
  - Return estimated as net interest (interest payable minus interest receivable).
  - Strengths: readily available; straightforward; applicable as alternative when balance sheet details are limited.
  - Drawbacks: omits equity costs; may understate capital costs for equity-funded entities.
- Option B2: Simplified WACC-based estimation (Task Team recommended).
  - Formula outline:
    - Cost of Capital = Nonfinancial Assets × Average Cost of Capital
    - Average Cost of Capital (nominal) = Ce × Nw/(Nw+L) + Cd × L/(Nw+L)
      - Ce = Cost of Equity (nominal net return to capital for nonmarket production per GFSM Update Proposed Recommendations 1.27).
      - Cd = Cost of Debt (ratio of total interest expenses to total stock of debt liabilities).
      - L = liabilities, excluding Equity liabilities.
      - Nw = Net Worth, including Equity liabilities (set to zero if negative).
  - Strengths: integrates debt and equity; aligns with 2025 SNA net return conceptualization; inputs obtainable from financial statements and consultation process.
  - Considerations: more calculations; omission of income tax adjustment for cost of debt to avoid complexity and because many public entities are tax-exempt.
- Option B3: Standardized rate of return for nonmarket producers.
  - Cost of Capital = Nonfinancial Assets × Rate of return for nonmarket producers.
  - Rate typically specified as “a nominal net rate of return of 2 percent plus a country’s inflation target.”
  - Strengths: straightforward and aligned with 2025 SNA notion of rate × asset value.
  - Drawbacks: does not account for full cost of debt; controversial to apply a nonmarket rate in a market test.
- Draft Task Team recommendations:
  - Depreciation and depletion: recommend use of financial-statement depreciation and depletion for market test practicality.
  - Return to capital: recommend adopting Option B2 (simplified WACC) in GFSM guidance; where balance sheet detail is unavailable, allow Option B1 (net interest) as alternative.
  - GFSAC October 2025: support for Task Team’s recommendation of Option B2.

### Issue C — Public utilities: classification challenges and options
Findings
- Public utilities (water, electricity, gas, energy, postal services, telecommunications) often operate as monopolies/oligopolies, face high fixed costs and economies of scale, and are subject to tariff regulation and public service obligations.
- These structural features do not automatically determine nonmarket status.
- Rate-of-return regulation can set tariffs to recover the cost of service (including depreciation and return to capital), equivalent to average-cost pricing and yielding zero economic profit.
Qualitative nonmarket indicators identified
- Additional barriers to entry (beyond natural ones).
- Tariff levels relative to average cost.
- Profitability profile.
Options for treatment
- Option C1: Maintain current GFSM 2014 guidance (simplicity; potential lack of clarity).
- Option C2: Provide more detailed guidance clarifying that monopoly/oligopoly status, tariff regulation, or public service obligations are not, by themselves, sufficient to classify units as nonmarket; evaluate under same market test consistent with choice in Issue A.
  - Draft Task Team recommendation: adopt Option C2.
  - GFSAC October 2025: broad support for Option C2.
- Option C3: Apply secondary indicators (from Option A3) to utilities with a decision rule: if two out of three secondary indicators are “yes”, classify as nonmarket. Advantages: objectivity; Drawbacks: data burden and potential incomplete capture of utilities’ specificities.

### Issue D — Regulatory agencies: sectoral classification (options and recommendation)
- Background: GFSM 2014 discusses market regulatory agencies and mixed functions; MGDD 2022 distinguishes market regulatory agencies from market regulatory bodies.
- Option D1: Maintain current GFSM 2014 guidance.
- Option D2: Simple decision tree to distinguish primary functions and treat subsidy distribution separately from regulatory function; if main activity is subsidy distribution, indicate nonmarket status.
  - Draft recommendation: adopt Option D2.
  - GFSAC October 2025: broad support for Option D2.
- Option D3: Agricultural-Parametrized decision tree replacing sector lists with decision rules.

### Issue E — Regulatory agencies: market test for mixed-function agencies
- Background:
  - GFSM 2014 advises splitting market operations and subsidy functions where possible.
  - MGDD 2022 proposes an 80 percent cost rule if separation infeasible: classify as market producer only if at least 80 percent of costs are directly linked to market activities.
- Option E1: Apply same criteria as for other public corporations; attempt separation; if infeasible, apply standard sales-to-cost test.
  - Draft recommendation: adopt Option E1.
  - GFSAC October 2025: support for Option E1.
- Option E2: Adopt MGDD 2022 80 percent cost rule (alignment with European practice); drawbacks include divergence from standard sales-to-cost test and additional accounting separation requirements.

### Issue F — Regulatory agencies: fees, licenses, and revenue in the market test
Findings
- 2025 SNA reclassifies compulsory nontransferable license fees as taxes; such revenues should be recorded as GFSM code 1145 and excluded from sales numerator in the market test.
- Tax collection by nongovernment units should be rerouted through government accounts; tax collection per se does not determine classification.
- Ambiguity when agencies have authority to impose taxes and set rates; may imply governmental attribution and affect classification.
Options
- Option F1: Maintain current GFSM 2014 guidance (does not reflect 2025 SNA reclassification).
- Option F2: Exclude compulsory nontransferable license revenues from the market test numerator; amend GFSM 2014 paragraphs to reflect 2025 SNA treatment and exclude taxes from sales-to-cost ratio.
  - Advantage: preserves uniformity and aligns with 2025 SNA; Limitation: only viable if 80 percent cost criterion is not adopted for Issue E.
- Option F3: Classify agencies with tax-imposing powers as government (creates clear rule aligned with tax attribution principles but creates an exception within market test framework).
Draft recommendations and GFSAC views
- Task Team recommends presenting Options F2 and F3 for consideration.
- GFSAC October 2025: divergence of opinions between merits of F2 and F3.

### Issue G — Regulatory agencies as regional or international institutions (residence and sectoral classification)
Options
- Option G1: Maintain current GFSM 2014 guidance.
- Option G2: Incorporate MGDD 2022 approach: if resident and market producer → public nonfinancial corporation of that country; if nonmarket → general government but separate inventory assets as quasi-corporation or notional resident unit.
- Option G3: Follow BPM7 approach — change residence classification depending on market test: market producer → public nonfinancial corporation; nonmarket → international organization and nonresident.
  - Task Team recommendation: adopt Option G3 and add guidance to Appendix 5 on market test effects on sectoral classification and residence.
  - GFSAC October 2025: support for Option G3.
Trade-offs
- Option G3 benefits: harmonizes standards, simplifies treatment of inventories and stocks, and captures residency dynamics; drawbacks: could inflate trade statistics and require significant revisions.

### Issue H — Nonprofit institutions (NPIs) controlled by public corporations
Options
- Option H1: Maintain current GFSM 2014 guidance.
  - Task Team recommendation: adopt Option H1.
  - GFSAC October 2025: divergence of opinions between H1 and H2.
- Option H2: Provide more detailed guidance: assess institutional unit status, if controlled classify as public unit, then apply market test to determine public corporation vs general government classification.
  - Advantage: clarifies that funding source (public corporation) does not preclude classification as a public unit.

### Key numeric thresholds and exact metrics proposed
- Sales-to-cost ratio threshold: 50 percent.
- Minimum analysis period for sales-to-cost ratio: three consecutive years.
- Debt/equity threshold referenced: 150 percent (Debt/Equity > 150 percent in combination with government-guaranteed or government-creditor debt > 50 percent).
- Government transfers dependency threshold: 50 percent (primary indicators and secondary indicators use 50 percent thresholds; a ratio of government transfers to total revenue threshold of 0.5 is referenced).
- Use of three-year averages for several indicators (debt/equity government support, share of transfers, transfers relative to sum of costs).
- MGDD 2022 alternative for regulatory agencies: 80 percent cost rule (market producer if at least 80 percent of total costs are directly linked to market activities).
- Standardized rate in Option B3: “a nominal net rate of return of 2 percent plus a country’s inflation target.”

### Draft Task Team recommendations (summary)
- Issue A (market test): Consider adopting one of Options A2, A3, or A4 to provide more prescriptive guidance; all three increase clarity and consistency.
- Issue B (sum-of-costs and return to capital):
  - Use financial-statement depreciation and depletion for market test practicality.
  - Adopt Option B2 (simplified WACC) for return to capital; permit Option B1 (net interest) where balance sheet detail is unavailable.
- Issue C (public utilities): Adopt Option C2 (provide more detailed guidance clarifying that monopoly/regulation/public service obligations are not alone determinative).
- Issue D (regulatory agency classification): Adopt Option D2 (simple decision tree).
- Issue E (regulatory agencies with mixed functions): Adopt Option E1 (apply same criteria as other public corporations; attempt separation; otherwise apply standard market test).
- Issue F (fees and licenses): Present Options F2 and F3 for consideration (exclude compulsory license revenues from market test and/or classify agencies with tax-imposing powers as government).
- Issue G (regional/international agencies): Adopt Option G3 (change residence classification depending on market test).
- Issue H (NPIs controlled by public corporations): Maintain current GFSM 2014 guidance (Option H1).
- Cross-cutting: Questions for global consultation include choices among Options A1–A4, B1–B3, C1–C3, D1–D3, E1–E2, F1–F3, G1–G3, H1–H2 and prompts on national practice and data availability.

*Discussion Note (GFSM 2014 Update Consultation: March 2026) — GFS compilation Task Team (TT1).*

### 2.1 Boundary between government-controlled nonmarket producers

### 2.1 Boundary between government-controlled nonmarket producers engaged in nonfinancial activities (government units) and public nonfinancial corporations

### Summary and purpose
- The Discussion Note identifies that the existing guidance in the Government Finance Statistics Manual 2014 (GFSM 2014) on classifying government-controlled nonfinancial units as market or nonmarket producers can lead to uneven implementation and reduced comparability of fiscal data.
- The Discussion Note considers adopting a more prescriptive framework for the market test that incorporates both quantitative and qualitative criteria to strengthen alignment with the 2025 System of National Accounts (2025 SNA).
- Proposed refinements address: reclassification of compulsory licenses as taxes; the sum-of-costs approach; guidance on measuring the return to capital; incorporation of depletion and rent of nonproduced assets as production cost items.
- The Discussion Note flags sector-specific cases requiring further clarification: public utilities, regulatory agencies, and nonprofit institutions controlled by public corporations.

*Task Team Responsible: GFS compilation Task Team (TT1). Authors: Artur Santos, Geraldine Kyalo, Irina Dubinina, Katherine Mills, Bocar Ka.*

### Introduction: problem statement and context
- GFSM 2014 paragraphs 2.64–2.75 provide principles for distinguishing market and nonmarket producers but leave room for differing national interpretations, which can affect coverage of general government and fiscal analysis (risk exposure, sustainability).
- Changes in the 2025 SNA affect the market test by:
  - Expanding scope of costs relevant to the market test to include rent payable on the use of nonproduced assets, depletion, and an explicit treatment of the net return to capital used in production.
  - Reclassifying compulsory nontransferable licenses as taxes, which must be excluded from sales revenue used in the market test.
- Eurostat MGDD 2022 (section 1.2.4.) and some IMF technical assistance have introduced more prescriptive methods (decision trees, indicators) that combine quantitative and qualitative criteria.

### Issue A — Need for more prescriptive guidance on the market test
Findings and background
- GFSM 2014 and 2025 SNA provide conceptual guidance on “economically significant prices” but deliberately refrain from rigid numerical thresholds; GFSM 2014 notes public corporations’ sales are expected, on average, to be “at least half of the costs of production over a sustained multi-year period,” implicitly suggesting a 50 percent benchmark while preserving flexibility.
- Most countries applying a systematic classification use a market test assessing whether sales cover at least half of production costs over a sustained multi-year period.
- Open methodological questions include:
  - The appropriate length of the “sustained multi-year period.”
  - The metric to relate sales-to-cost ratios within that period (average of annual ratios versus consecutive-year rule).
- Manuals specify for sales receipts: measure before applying product taxes; exclude subsidies on products unless granted to any producer undertaking the same activity; exclude payments received from government unless they would be made to similar producers; exclude own-account production.
- Production costs coverage is addressed via the sum-of-costs approach (see section B in the Discussion Note).

Implications of 2025 SNA changes
- Inclusion of rent of nonproduced assets and depletion, and explicit net return to capital, directly affect quantitative assessment of prices and cost coverage.
- Reclassification of compulsory licenses as taxes requires excluding such revenues from sales used in the market test.

Complementary practice
- Eurostat MGDD 2022 offers a more prescriptive, rules-based approach including a quantitative market test plus qualitative criteria and a decision tree.
- Practical approaches used in technical assistance (Tanzer & Stokoe, 2025) introduce structured combinations of qualitative and quantitative indicators (government support, profitability, competition, price setting).

### Options identified to strengthen guidance (Options A1–A4)
Option A1: Maintain current GFSM 2014 guidance
- Preserve principles-based approach and flexibility to account for country-specific factors.
- Risk: continued cross-country and temporal discrepancies in classification outcomes.
- Note: Even if retained, updates would be necessary to reflect the 2025 SNA revised sum-of-costs approach, affecting units with significant rent or depletion costs.

Option A2: Decision-tree approach (more prescriptive quantitative and qualitative guidance)
- Adopt the MGDD 2022 decision tree and apply qualitative criteria before a quantitative test.
- Use a 50 percent threshold for the sales-to-cost ratio calculated over at least three consecutive years.
- Quantitative test subject to additional considerations:
  - Exclude results impacted by one-off or exceptional circumstances (e.g., due to the Covid pandemic).
  - Allow early reclassification if justified by reasonable expectations about future performance.
  - Use forecasts of sales-to-cost ratios for start-up entities until sufficient historical data exist.
- Benefits: enhances international comparability while preserving flexibility for special cases.
- Risks: over-reliance on a single indicator; potential to overlook factors such as degree of government pricing control, nature of goods/services, and entity influence on market price, supply, and demand.

Option A3: Indicator approach (primary and secondary indicators with thresholds)
- Build a descriptive framework around common primary and secondary indicators, decision rules, and numerical thresholds, drawing from MGDD 2022 and Tanzer’s approach.
- Primary Indicators:
  - Primary Indicator 1: Is the government the unit’s sole client? (yes or no)
  - Primary Indicator 2: Ratio of sales to production costs over the last three consecutive years.
  - Primary Indicators 3 (set):
    - Is more than 50 percent of the unit’s debt either government-guaranteed or provided directly by government as creditor, provided that total liabilities exceed 150 percent of equity (three-year average)? (yes or no)
    - Is more than 50 percent of revenue composed of government transfers (three-year average)? (yes or no)
    - Are transfers from government greater than 50 percent of the sum of remuneration of employees, use of goods and services, rent of nonproduced assets, depreciation, depletion, other taxes on production, and acquisitions of fixed assets (three-year average)? (yes or no)
- Secondary Indicators:
  - Profitability indicator: Negative equity? (yes or no)
  - Limited competition indicator: Are there prohibitive barriers to entry established by government or by legal restrictions? (yes or no)
  - Price control indicator: Are prices set directly by government, and are they defined at a level below the unit cost of production, including taxes? (yes or no)
- The approach can be operationalized via alternative rules for combining indicators (see Annex I referenced in the Discussion Note).
- Benefits: structured, operationally clear framework supporting transparency and consistency.
- Risks: reduced flexibility; potential oversimplification; increased data requirements and compilation burden.

Option A4: Hybrid decision-tree plus indicators
- Integrate elements of the indicator-based framework into a modified decision-tree approach.
- Retain initial qualitative steps of the decision tree, then apply the indicator-based framework (Primary Indicator 2, Primary Indicators 3, and Secondary Indicators) instead of immediately applying a standard sales-to-cost ratio test.
- Benefits: combines structured decision pathway with richer indicator assessment to reduce overreliance on a single threshold.

### Practical implementation considerations
- Any prescriptive framework must define:
  - Exact components of sales revenue and production costs to be included (e.g., exclusion of compulsory license revenues reclassified as taxes).
  - The appropriate multi-year analysis window (the Discussion Note cites three consecutive years in proposals).
  - Procedures to adjust for exceptional one-off events and for start-up entities using forecasts.
- Data and compilation burdens: Option A3 and Option A4 increase data requirements (multiple ratios, three-year averages, debt/equity assessments).
- Risk of misclassification if strict thresholds are applied without contextual assessment (market structure, pricing controls, essential service nature).

### Sector-specific clarifications flagged
- Public utilities: price regulation and economies of scale do not automatically imply prices are not economically significant; need clearer rules.
- Regulatory agencies: combined direct and indirect market interventions require clear treatment under the market test.
- Regional arrangements and agencies controlled by economic unions: sectoral classification and residency can depend on the market test outcome.
- Nonprofit institutions controlled by public corporations: raise specific sectoral classification questions when operating as nonmarket producers.

### Key numeric thresholds and exact metrics proposed in the Discussion Note
- Implicit and proposed numerical threshold for sales-to-cost ratio: 50 percent.
- Proposed minimum analysis period for sales-to-cost ratio: three consecutive years.
- Debt/equity threshold referenced: 150 percent (Debt/Equity > 150 percent in combination with government-guaranteed or government-creditor debt > 50 percent).
- Government transfers dependency threshold referenced: 50 percent (primary indicators and secondary indicators use 50 percent thresholds; a ratio of government transfers to total revenue threshold of 0.5 is referenced).
- Use of three-year averages for several indicators (debt/equity government support, share of transfers, transfers relative to sum of costs).

*Discussion Note authors and Task Team (GFS compilation Task Team (TT1)).*

### 20.      Option A4 offers the advantage of combining the intuitive qualitative sequencing of a decision

### discussion-note-21-boundary-between-government-and-public-nonfinancial-corporations - 20.      Option A4 offers the advantage of combining the intuitive qualitative sequencing of a decision

### Option A4 and Draft Recommendations (Market test guidance)
- Option A4 combines a qualitative decision-tree sequencing with a quantitative indicator approach; Annex II presents decision-tree versions of the indicator combinations considered under Option A3.
- Decision trees in Annex II extend the decision tree in Figure 2 to answer: Does the unit qualify as a market producer based on the indicator approach?
- Advantages of Option A4:
  - Combines intuitive qualitative sequencing with a detailed quantitative assessment.
  - Enhances clarity by first assessing whether the entity is a dedicated provider of ancillary services to government before evaluating indicators.
- Drawbacks of Option A4:
  - More complex than Options A2 or A3.
  - Increases data requirements and operational burden on compilers.
- Draft Task Team recommendation:
  - The Task Team considers current guidance inadequate and proposes Options A2, A3, or A4 as recommended alternatives. All three:
    - Provide more prescriptive guidance.
    - Offer a clearer and more structured framework for applying the market test.
    - Promote greater consistency and transparency in sector classification.
- GFS Advisory Committee (GFSAC) October 2025 meeting:
  - Diversity of views; substantive support for Options A2 and A3.
  - Divergence on whether a decision-tree (A2) or indicator-based (A3) approach is preferable.
  - Option A4 developed following the meeting as a hybrid to address concerns on both sides.

*GFSM 2014 Update Consultation: March 2026 — Draft recommendations on market test approaches and Option A4.*

### Issue B — Sum of Costs approach: background and components
- 2025 SNA uses the sum-of-costs approach to value production for own final use (when no reliable market prices exist) and nonmarket production by government and NPISHs when output is free or at prices not economically significant.
- 2008 SNA components included: intermediate consumption, compensation of employees, other taxes less subsidies on production, consumption of fixed capital, and a return to capital employed in production.
- 2025 SNA expands the sum-of-costs by introducing:
  - Depletion (of natural resources used in production).
  - Rent payable for use of nonproduced nonfinancial assets.
  - Some adjustments to cost components.
- The production cost components considered in the market test analyzed in the 2025 SNA include:
  - Remuneration of employees (new term for compensation of employees in 2025 SNA).
  - Use of goods and services.
  - Rent of nonproduced assets.
  - Depreciation (new term for consumption of fixed capital in 2025 SNA).
  - When applicable, depletion.
  - Other taxes on production (without deduction of subsidies).
  - A return to capital employed in production.
- GFSM Research Project 1.27 examined the application of the sum-of-costs approach and recommended exploring implications for the market test.

### Rent (cost component)
- Definition (GFSM 2014 paragraph 5.122): “Rent is the income received by the owner of a nonproduced natural resource or other nonproduced nonfinancial asset (the lessor) for making that resource or nonproduced asset available to another institutional unit (the lessee) for use in production.”
- 2025 SNA broadens rent to include payments for access to a wider range of nonproduced assets beyond natural resources (paragraph 30.33), including:
  - Land rentals.
  - Rentals of mineral and energy resources.
  - Use of nonproduced intangible assets such as contracts, licenses, goodwill, and marketing assets (brand reputation and customer loyalty associated with corporate patents or trademarks).
- Implication: Rent should be included as a cost component in the market test under the GFSM update (GFSM Update PR 1.27).

### Depletion and Depreciation
- Depletion:
  - Refers to reduction in the value of nonproduced natural assets (e.g., mineral reserves, energy resources, exploited forests) from productive use.
  - 2025 SNA incorporates depletion as a new component in the sum-of-costs, raising overall costs and potentially making it harder for public entities in extractive sectors to pass the quantitative market test, possibly leading to reclassification to general government.
- Depreciation:
  - 2025 SNA renames consumption of fixed capital as “depreciation”.
  - Distinguish statistical depreciation (based on current exchange values, reflecting economic loss) from accounting depreciation (historical-cost-based accounting practice).
  - Practical note: compilers generally rely on corporate financial statements for depreciation; recalculating statistical depreciation independently could be excessively complex.

### Return to Capital: conceptual approaches and issues
- SNA Issue Note A.8: Using capital assets in production adds value beyond depreciation; financing implies costs (interest or opportunity cost of equity).
- GFSM 2014 lists return to capital as a cost but lacks calculation details.
- MGDD 2022 (section 1.2.4) uses net interest expense as a proxy for return to capital.
- 2025 SNA (paragraph A5.25) defines net return to capital as rate of return multiplied by the value of nonfinancial assets used in production — shifting from financial-statement expense data to estimating a rate of return applied to nonfinancial asset values.
- Methodological tension:
  - Applying a nonmarket producer rate in a market test may be inconsistent because market test aims to reflect market revenues and costs.
  - Corporate finance literature proposes WACC as conceptually robust, incorporating both equity and debt components.
- WACC:
  - Represents average return required by debt and equity holders, weighted by capital structure.
  - Academic methods (e.g., CAPM) require inputs often unavailable for government-controlled or nonpublicly traded entities.
  - The discussion note proposes a practical and accessible method preserving conceptual clarity without advanced calculations.

### Options identified to address return-to-capital estimation (Options B1, B2, B3)
- All options incorporate GFSM Update Notice of Decision 1.27 recommendations: include depletion and rent of nonproduced assets as production cost items and add a GFSM footnote noting financial-statement depreciation and depletion may differ from statistical measures in 2025 SNA and GFSM; but financial-statement values may be used for market test practicality.

- Option B1: Net interest as return to capital
  - Estimate return to capital as net interest charge (interest payable minus interest receivable) based on MGDD 2022 (section 1.2.4).
  - Strengths:
    - Plausible proxy for financing costs.
    - Readily available information; straightforward to implement.
    - Facilitates application to own-account capital formation (net interest related to own-account capital formation could be reclassified as a net transaction in nonfinancial assets).
  - Drawbacks:
    - Conceptually limited: captures only debt-related costs, ignores equity financing.
    - May understate total capital costs for entities funded largely by equity, reducing comparability across entities with varying capital structures.

- Option B2: Simplified WACC-based estimation (recommended by Task Team)
  - Proposed simplified formula:
    - Cost of Capital = Nonfinancial Assets × Average Cost of Capital
    - Average Cost of Capital (nominal) = Ce × Nw/(Nw+L) + Cd × L/(Nw+L)
      - Ce (Cost of Equity): Nominal net return to capital for nonmarket production, as estimated based on GFSM Update Proposed Recommendations 1.27.
      - Cd (Cost of Debt): Ratio of total interest expenses to total stock of debt liabilities.
      - L = liabilities, excluding Equity liabilities.
      - Nw = Net Worth, including Equity liabilities (set to zero if negative).
  - Alignment with 2025 SNA: net return to capital = rate of return × value of nonfinancial assets.
  - Strengths:
    - Integrates both debt and equity financing; conceptually coherent with 2025 SNA.
    - Data required (cost of debt, capital structure weights) are obtainable from financial statements; cost of equity derived from global consultation process.
  - Considerations:
    - More calculations required but inputs are generally available.
    - The proposed WACC approach omits an income tax adjustment for cost of debt to avoid complexity and because many public entities are tax-exempt or tax arrangements may render such adjustments immaterial.

- Option B3: Standardized rate of return for nonmarket producers
  - Cost of Capital = Nonfinancial Assets × Rate of return for nonmarket producers.
  - The rate is the nominal net return to capital for nonmarket production per GFSM Update Proposed Recommendations 1.27; typically, “a nominal net rate of return of 2 percent plus a country’s inflation target.”
  - Strengths:
    - Conceptually aligned with 2025 SNA and straightforward to apply since nominal rate is defined in GFSM Update Proposed Recommendations 1.27.
  - Drawbacks:
    - Does not account for full cost of debt paid.
    - Methodological controversy: applying a nonmarket rate to a market test where costs should reflect market conditions.

### Draft Task Team recommendations on depreciation/depletion and return to capital
- Depreciation and depletion:
  - Recommend use of depreciation and depletion from financial statements for applying the market test.
  - Rationale: compilers generally rely on corporate financial statements; independent recalculation of statistical depreciation would be overly complex.
  - Clarification to be included in updated GFSM recommending use of financial statement depreciation and depletion for market test purposes.

- Return to capital:
  - Recommend adopting Option B2 (simplified WACC framework) in GFSM guidance.
    - Combines quantitative market test with qualitative criteria.
    - Uses a simplified WACC to estimate return to capital, aligning with 2025 SNA changes (return = rate × value of nonfinancial assets and using nominal net return to capital for nonmarket production as a proxy for cost of equity).
    - Includes both debt and equity components for a more complete representation of capital costs.
    - Inputs can be sourced from financial statements: cost of debt and capital structure from entity accounts, cost of equity from the global consultation process linked to the GFSM update.
  - Contingency:
    - Where an entity’s balance sheet is unavailable or lacks detail, Option B1 (net interest) may be used as an alternative because it requires less financial information.
- GFSAC October 2025: there was support for the Task Team’s recommendation of Option B2.

*GFSM 2014 Update Consultation: March 2026 — Task Team recommendations on sum-of-costs, depreciation/depletion usage, and return-to-capital estimation (Options B1–B3).* 

### Issue C — Public utility companies: background and implications for market classification
- Public utilities examples: water, electricity, gas, energy, postal services, telecommunications.
- Challenges:
  - Often operate as monopolies or oligopolies with high fixed costs and economies of scale.
  - Subject to tariff regulation and public service obligations affecting revenue and cost structure.
  - These structural features do not, by themselves, determine nonmarket status.
- Relevant guidance:
  - Consistent with paragraph 30.26 of 2025 SNA and GFSM 2014 paragraph 2.66: lack of autonomy to set prices and output levels may lead some countries to classify enterprises as nonmarket producers, even when regulated tariffs cover production costs including return to capital (which could be economically significant).
- Key principles:
  - Existence of a natural monopoly does not automatically imply non-economically significant prices; there are public and private natural monopolies.
  - Government control affects classification as public vs private but should not influence assessment of whether prices are economically significant.
  - Classification should be based on assessment of economically significant prices distinguishing market from nonmarket producers.
- Regulatory context:
  - Unregulated monopolies set prices above marginal cost, generating deadweight loss; governments often regulate prices.
  - Rate-of-return regulation commonly sets tariffs to recover cost of service (including depreciation, taxes, and return to capital), akin to sum-of-costs valuation.
  - Rate-of-return regulation corresponds to a price equal to average cost, yielding zero economic profit (accounting profit positive because it includes return to capital).

*GFSM 2014 Update Consultation: March 2026 — Discussion of public utilities and implications for market/nonmarket classification.*

### 49.      For public utilities, qualitative features that serve as relevant nonmarket indicators include:

### discussion-note-21-boundary-between-government-and-public-nonfinancial-corporations - 49.      For public utilities, qualitative features that serve as relevant nonmarket indicators include:

### Public utilities: qualitative nonmarket indicators and options
- Qualitative features identified as relevant nonmarket indicators:
  - additional barriers to entry (beyond natural ones)
  - tariff levels relative to average cost
  - profitability profile
- Relation to Option A3:
  - "These aspects are already covered under the secondary indicators of Option A3. Thus, if that option is adopted, the application to utilities would be directly addressed."
  - "If another option prevails to address Issue A, it may be necessary to consider applying these secondary indicators specifically to public utilities."
- Option C1: Maintain the current GFSM 2014 guidance.
  - Advantage: simplicity; no additional guidance or clarification required.
  - Drawback: potential lack of clarity where pricing is shaped by complex regulatory or political dynamics.
- Option C2: Provide more detailed guidance on the classification of public utility companies.
  - Clarification: utility companies operating as a monopoly or oligopoly, subject to tariff regulation, and fulfilling public service obligations are not, in themselves, sufficient to classify them as nonmarket producers, even if the producer lacks complete autonomy to set prices and output levels.
  - Emphasis: evaluate these enterprises under the same classification rules and subject them to the same market test as other government-controlled entities, consistent with the option chosen in section A.
  - Advantage: recognizes distinctive operating conditions of public utilities and could improve classification accuracy across countries.
  - Disadvantage: introducing further guidance might not lead to meaningful changes in classification outcomes or add value beyond existing criteria.
- Option C3: Apply secondary indicators to public utilities (if Option A3 or A4 is not adopted).
  - Three secondary indicators proposed:
    - Profitability Indicator: Is equity negative? (yes/no)
    - Limited Competition Indicator: Are there entry barriers defined by government or legal restrictions? (yes/no)
    - Government Price Control Indicator: Are government-set prices below unit cost of production, including taxes? (yes/no)
  - Decision Rule: "If two out of three indicators are “yes”, then classify the unit as nonmarket."
  - Advantage: greater objectivity; measurable evidence reinforces international comparability and replicability.
  - Drawback: requires greater analytical effort to calculate indicators for each enterprise; may not fully capture all specificities of public utilities.
  - Note: "Although the transport and education sectors may operate under public service obligations and be subject to strong regulation, they almost always face effective competition from private providers. If Option C3 is selected, its indicator-based approach can also be applied directly to these sectors."
- Draft recommendation:
  - "The Task Team recommends updating GFSM guidance by adopting Option C2."
  - Rationale: regulatory constraints and public service obligations complicate standard market tests; expanded guidance needed to clarify how these characteristics should be taken into account in sector classification decisions.
  - Meeting support: "At the October 2025 GFSAC meeting there was broad support for the Task Team’s recommendation of Option C2."

### Issue D — Regulatory agencies: sectoral classification (background and options)
- Background:
  - GFSM 2014 describes "market regulatory agencies" distinguishing direct and indirect market intervention and mentions possibility that such agencies distribute subsidies as well as regulate market activities (paragraphs 2.156–2.159).
  - MGDD 2022 (section 1.4.) more narrowly defines market regulatory agencies and separates:
    - "Market regulatory agencies" (section 1.4.): direct intervention (buying, selling, and stocking goods).
    - "Market regulatory bodies" (section 1.6.5.): indirect intervention (rules, licensing, supervision, sanctions).
  - Convention: financial regulatory bodies are part of the financial auxiliary corporations sector, not government; rationale could be extended to agencies regulating agricultural products, monopolistic markets, and natural resources.
- Option D1: Maintain current GFSM 2014 guidance.
  - Advantage: preservation of established understanding and conceptual stability; avoids changes in data compilation.
  - Drawback: preserves existing uncertainty and inconsistent national interpretation.
- Option D2: Simple decision tree.
  - Proposal: supplement GFSM 2014 paragraph 2.158 with a decision tree; add textual guidance that distribution of subsidies should be treated separately from regulatory function; assess whether unit operates primarily as market regulator or as distributor of subsidies.
  - Indicator: If main activity is distribution of subsidies, this constitutes an indication that the unit is nonmarket.
  - Advantage: enhances clarity and provides logical sequence for classification.
  - Drawbacks: may require reclassification and data adjustments; may leave room for different interpretations in hybrid cases.
- Option D3: Agricultural-Parametrized decision tree.
  - Proposal: remove from GFSM 2014 paragraph 2.156 the list of sectors and replace paragraph 2.158 bullet points with a decision tree; retain textual clarification distinguishing subsidy distribution from regulatory functions.
- Draft recommendation:
  - "The Task Team recommends updating GFSM guidance by adopting Option D2."
  - Rationale: amending GFSM 2014 paragraphs 2.156–2.159 to provide clearer decision rules for classifying agencies involved in market and nonmarket activities, while not binding classification to specific sectors of operation.
  - Meeting support: "At the October 2025 GFSAC meeting there was broad support for the Task Team’s recommendation of introducing a simple decision tree - Option D2."

### Issue E — Regulatory agencies: market test (background and options)
- Background:
  - ESA 2010 (paragraph 20.53) and MGDD 2022 (section 1.4.) suggest an alternative approach for regulatory agencies focused on the "main activity" (subsidy distribution or market operations).
  - GFSM 2014 paragraph 2.159 recognizes agricultural regulatory agencies combining subsidy distribution with market operations; first recommendation is to attempt a split into two segments and classify separately: market operations as quasi-corporation (public nonfinancial corporation sector) and subsidy functions as government.
  - If separation not feasible, MGDD 2022 proposes an 80 percent cost rule: classify as market producer only if at least 80 percent of total costs are directly linked to market activities; if subsidy-related costs exceed 20 percent, classify within general government.
  - Note: the composition-of-costs approach may diverge from the standard sales-to-cost test; a profitable agency might be classified within government if subsidy-related costs exceed 20 percent.
- Option E1: Apply the same criteria used for other public corporations.
  - Process: attempt separation; if not possible, apply standard market test (sales-to-cost) to the agency as a whole.
  - Advantage: methodological consistency and international comparability; preserves sales-to-cost test as anchor.
  - Drawback: without an alternative like the 80 percent rule, the standard test may understate the operational relevance of subsidy-distribution activities.
- Option E2: Adopt the MGDD 2022 criterion (80 percent cost rule).
  - Process: attempt separation; if not feasible, apply 80 percent test (market producer if at least 80 percent of total costs are directly linked to market activities; otherwise nonmarket).
  - Advantages: alignment with European practice.
  - Disadvantages:
    - introduces a different test than the standard sales-to-cost criterion, potentially creating international inconsistencies;
    - may classify as government certain agencies that operate profitably because subsidy costs exceed 20 percent;
    - requires additional accounting separation of direct market costs from subsidy costs, not always feasible and difficult to compare across countries.
- Draft recommendation:
  - "The Task Team recommends updating GFSM guidance by adopting Option E1."
  - Rationale: preserves uniformity in application of the market test, enhances comparability, maintains sales-to-cost test as anchor, and recommends institutional separation of activities wherever possible.
  - Meeting support: "At the October 2025 GFSAC meeting there was support for the Task Team’s recommendation of Option E1."

### Issue F — Regulatory agencies: fees, licenses, and revenue treatment in the market test
- Background:
  - 2025 SNA (consistent with SNA Update Guidance Note WS.14 and GFSM Update Proposed Recommendations 1.24) reclassifies compulsory nontransferable license fees as taxes.
  - When such revenues are collected by regulatory agencies, they must be recorded in government accounts as "Other taxes on the use of goods and on permission to use goods or perform activities" (GFSM code 1145).
  - Tax collection by nongovernment units: possible for a nongovernment unit to collect taxes on behalf of government (GFSM 2014 paragraphs 5.2, 5.33-5.40 and 2025 SNA paragraph 30.170), but transactions should be rerouted through government accounts; tax collection alone does not determine classification as government.
  - Ambiguity arises when the agency has authority to impose the tax and set the rate: such tax attribution is inherently governmental and could suggest classification in general government if exercised by the agency itself.
  - Classification may depend on whether taxes constitute a substantial share of the agency’s revenue.
  - From the market test perspective: "the concept of economically significant prices is not valid when taxes are included in revenue."
  - Proposed consistent solution: exclude taxes from the numerator of sales in the sales-to-cost ratio used in the market test (see Issue A).
  - Caveat: If the 80 percent cost criterion for regulatory agencies is adopted, excluding taxes from sales would no longer apply since that test does not rely on the sales-to-cost ratio.
- Option F1: Maintain current GFSM 2014 guidance.
  - Advantage: preserves established procedures, methodological stability, and minimizes transition costs.
  - Drawback: does not adjust the market test to the reclassification of compulsory licenses as taxes under 2025 SNA; may compromise international comparability and would no longer align with 2025 SNA.
- Option F2: Exclude compulsory nontransferable license revenues from the market test.
  - Proposal:
    - amend GFSM 2014 (paragraph 2.73) to explicitly exclude tax revenues (including government compensation for tax collection) from the sales numerator in the market test;
    - revise GFSM 2014 (paragraphs 2.156–2.259) to clarify that compulsory nontransferable licenses are taxes, with reference to new GFSM paragraph 3.28 (incorporating 2025 SNA, paragraph 30.170), reinforcing that taxes must be excluded from the sales-to-cost ratio.
  - Advantage: preserves uniformity of treatment; agencies would be classified as nonmarket producers when tax revenues make up a substantial share of income; conceptually consistent with 2025 SNA.
  - Limitation: only viable if the option for Issue E does not adopt the 80 percent cost criterion.
- Option F3: Classify agencies with tax-imposing powers as government.
  - Proposal:
    - in addition to amending GFSM 2014 (paragraph 2.73) to exclude tax revenues from the market test numerator, add a new provision establishing, by convention, that agencies with authority to impose taxes and set rates should be classified as government units.

*GFSM 2014 Update Consultation: March 2026 — GFSM 2014 update*

### 88.      The advantage is the creation of a clear rule aligned with the principles of tax attribution.

### discussion-note-21-boundary-between-government-and-public-nonfinancial-corporations - 88.      The advantage is the creation of a clear rule aligned with the principles of tax attribution.

### Regulatory agencies — Fees, licenses, and revenue (Issue F)
- Paragraph 88: Option described creates a clear rule aligned with principles of tax attribution; compatible with prior choices; drawback is a differentiated classification rule specifically for regulatory agencies, creating an exception within the market test framework.
- Draft recommendations:
  - Paragraph 89: The Task Team recommends presenting Options F2 and F3 for consideration. Both options adjust the sector classification of regulatory agencies in response to the reclassification of compulsory nontransferable licenses as taxes under the 2025 SNA.
  - Paragraph 90: At the October 2025 GFSAC meeting there was a divergence of opinions between the merits of Option F2 (a principles-based approach) and Option F3 ( a rules-based apporach).

### Regulatory agencies — Classification as regional or international institutions (Issue G)
- Background (paragraphs 91–95):
  - Some regulatory agencies perform market regulation and subsidy management and may be under economic ownership of economic unions; key factor is whether the agency follows common policies of the economic union.
  - GFSM 2014 (paragraph 2.158 and paragraph A5.2) states regulatory agencies that qualify as international or regional organizations should not be included in member-country statistics but rather in regional data, provided they qualify as institutional units.
  - MGDD 2022 (section 1.4) treats agricultural regulatory agencies on behalf of the European Union as:
    - If market producer: classified as a public nonfinancial corporation.
    - If nonmarket producer: classified within the general government sector, but inventory assets held by the agency would be separated because economic ownership belongs to the union; the unit holding inventories could be treated as (i) a quasi-corporation classified as a public nonfinancial corporation; or (ii) a notional resident unit (resident unit while owner is the RoW).
  - GFSM 2014 paragraph 2.13 and paragraph 7.61 note limits on applying notional resident unit concepts to inventories (movable goods rather than immovable assets).
  - BPM7 (paragraphs 4.184–4.185, 4.188, 4.198) gives primacy to the market test for institutional classification and residence, recommending classification as international organization for regional organizations exercising government functions and financed mainly by intergovernmental transfers; if market producer, classify as resident public nonfinancial corporation in economy of predominant center of economic interest, or apportion data among member countries.
- Options identified (paragraphs 96–101):
  - Option G1: Maintain current GFSM 2014 guidance.
    - Benefit: preserves room for national interpretation; tailored approaches; does not substantially undermine comparability.
    - Drawback: does not resolve complexity for regional regulatory agencies.
  - Option G2: Incorporate the MGDD 2022 approach.
    - Guidance: If agency resident in Country A and market producer → public nonfinancial corporation of Country A; if nonmarket producer → general government of Country A but separate inventory assets with unit holding inventories treated as (i) quasi-corporation or (ii) notional resident unit (owner is RoW).
    - Benefit: avoids artificial trade flows; aligns with EU practice.
    - Drawback: complexity of separating inventories; notional resident units for inventories may be conceptually inconsistent and pose implementation challenges.
  - Option G3: Follow BPM7 approach — change residence classification depending on market test.
    - Guidance: If market producer → public nonfinancial corporation; if nonmarket producer → international organization and nonresident of domestic economy.
    - Benefits (paragraph 101):
      - Harmonizes international statistical standards and strengthens consistency between GFSM and BPM7.
      - Provides clearer and less complex framework by avoiding methodological separation of inventory assets and related stocks/flows within same institutional unit.
      - Avoids challenges to vertical integration of macroeconomic statistics (e.g., identifying liabilities, recording transactions and other economic flows).
      - More accurately captures residency dynamics of units operating under common economic union policies.
    - Drawbacks (paragraph 102):
      - Exports and imports would need to be recorded every time the agency buys or sells inventories, potentially inflating trade flows statistics.
      - Could require significant revisions to current statistics and reclassification of units, implying additional workload.
- Draft recommendations (paragraphs 103–104):
  - Paragraph 103: The Task Team recommends updating GFSM guidance by adopting Option G3, adding guidance to Appendix 5 of GFSM 2014 on how the market test can affect both sectoral classification and residence; could be presented as a special case of public sector institutional unit decision tree (GFSM 2014 paragraph 2.124).
  - Paragraph 104: At the October 2025 GFSAC meeting there was support for the Task Team’s recommendation of Option G3.

### NPIs controlled by public corporations (Issue H)
- Background (paragraphs 105–108):
  - In some countries public corporations establish nonprofit institutions (NPIs) that operate in quasi-fiscal activities or have nonmarket purposes and receive most resources through transfers from the public corporations that created them.
  - Classification questions arise because funding comes primarily from public corporations rather than direct general government budgetary transfers — should such NPIs be extrabudgetary units of government or remain outside government because financed by a public corporation?
  - GFSM 2014 (paragraph 2.36) and decision tree for public units (2025 SNA paragraph 30.35; GFSM 2014 paragraph 2.124) apply.
  - If foundations qualify as institutional units and are controlled by another public unit, they are classified as public units. Market producers → public corporations sector; nonmarket producers → general government.
  - GFSM 2014 (paragraph 2.37, paragraph 2.128, paragraph 2.82) provides descriptions and options including classification as extrabudgetary units.
- Options identified (paragraphs 109–111):
  - Option H1: Maintain the current GFSM 2014 guidance (paragraph 109).
  - Option H2: Provide more detailed guidance on NPIs controlled by public corporations (paragraph 110), with the proposed procedure:
    - Step 1: Assess whether these foundations qualify as institutional units.
    - Step 2: If they do, and are controlled by another public unit, classify them as public units (considered controlled by government).
    - Step 3: Determine whether they are market or nonmarket producers (apply market test described in section A of the Note): market producers → classify within public corporations sector; nonmarket producers → classify within general government sector.
  - Advantages of Option H2: eliminates ambiguity between funding source and sectoral classification; clarifies that a public corporation (market producer) can create and control a government institution (nonmarket producer).
  - Disadvantage of Option H2: additional guidance may not change classification outcomes or add much value beyond existing criteria.
- Draft recommendations (paragraphs 112–113):
  - Paragraph 112: The Task Team recommends maintaining the current GFSM 2014 guidance by adopting Option H1.
  - Paragraph 113: At the October 2025 GFSAC meeting there was a divergence of opinions between the merits of Option H1 (no change) and Option H2 (additional clarification).

### Cross-cutting procedural and consultation notes
- Paragraph 114: An earlier version of the discussion note was discussed at the GFSAC meeting of October 2025; the task team recommendations are consistent with the preliminary views of most GFSAC members.
- Questions for global consultation (selected issues and prompts, Paragraphs following 114):
  - Issue A: Choice among Option A1, Option A2, Option A3, Option A4; reliance on market test versus qualitative criteria; appropriateness of the 50 percent threshold.
  - Issue B: Choice among Option B1, Option B2, Option B3; data availability for weighted cost to capital; inclusion of depletion and/or rent.
  - Issue C: Choice among Option C1, Option C2, Option C3; classification challenges of public utility companies.
  - Issue D–H: Choices among specified options (D1–D3, E1–E2, F1–F3, G1–G3, H1–H2) with prompts about national practice and ability to separately identify functions (e.g., subsidy distribution vs market activities).

---


_Source: https://www.imf.org/-/media/files/data/statistics/gfsm/global-consultations/discussion-note-21-boundary-between-government-and-public-nonfinancial-corporations.pdf_