Skip to content
Provisional draft release

No canonical PSII Core v2 release is selected. Everything shown is provisional draft data and must not be cited as a published PSII score.

Displayed: 2026.1-draft

Methodology

How PSII Core v2 is built

PSII Core v2 is an annual, versioned, cross-national structural index of how resilient a country is to undue private-sector influence in strategic and policy environments. It is a structural-conditions index, not a measure of influence events.

Score direction. Higher score = more resilient, less structurally susceptible to undue private-sector influence. Scores are displayed to 3 decimals, which is the precision the release publishes and not a claim of accuracy to that digit.

Version separation

PSII Published v1

The 2023 published edition is archival and tied to the original publication. It is not merged into the annual Core v2 rankings.

PSII Core v2

The annual structural index uses the revised v2 methodology. It is resilience-oriented and not numerically identical to the chapter table.

Score formula

Each dimension is the equal-weighted arithmetic mean of its normalized indicators. The overall Core v2 score applies theory-led dimension weights.

PSII = 0.50 · Governance + 0.35 · Economic + 0.15 · Information

Governance and elite capture

50%

Quality of institutions and resistance to elite capture. Equal-weighted mean of 8 normalized indicators.

  • Control of corruptionWorld Bank WGI · wgi_control_of_corruption
  • Rule of lawWorld Bank WGI · wgi_rule_of_law
  • Regulatory qualityWorld Bank WGI · wgi_regulatory_quality
  • Government effectivenessWorld Bank WGI · wgi_government_effectiveness
  • Voice and accountabilityWorld Bank WGI · wgi_voice_accountability
  • Egalitarian democracyV-Dem · vdem_egalitarian_democracy
  • Political corruption (inverted)V-Dem · vdem_political_corruption
  • Corruption perceptionsTransparency International · ti_cpi

Economic structure and external dependence

35%

Openness and structural diversification of the economy. Equal-weighted mean of 3 normalized indicators.

  • FDI openness (inverted)OECD · oecd_fdi_restrictiveness
  • Export diversification (inverted)UNCTAD · unctad_export_product_concentration
  • Import diversification (inverted)UNCTAD · unctad_import_product_concentration

Information environment and media independence

15%

Independence and plurality of the information space. Equal-weighted mean of 2 normalized indicators.

  • Freedom of expressionV-Dem · vdem_freedom_expression
  • Press freedomRSF · rsf_press_freedom

Normalization

  • Provider 0 to 100 scales (WGI, CPI, RSF) are divided by 100.
  • Provider 0 to 1 indices (V-Dem) are used as published.
  • Vulnerability-oriented scales (political corruption, FDI restrictiveness, trade concentration) are inverted so that every indicator points the same way.
  • Each dimension is the equal-weighted mean of its normalized indicators.
  • No year-relative percentile scaling and no imputation are used for the public score.

Release discipline

A canonical annual release requires every core indicator for the same reference year, one frozen source vintage per provider, and explicit release metadata.

No hidden imputation is allowed. If a country-year is missing any core indicator, its score status is incomplete, its scores stay null, and the missing indicators are listed.

Displayed release 2026.1-draft · methodology PSII Core v2.1-draft · coverage 16 of 16 country-years

Limitations that stay public

Structural, not event-level

A low score flags structural susceptibility. It does not prove that any company, oligarchic network, or foreign actor influenced a specific decision, crisis, or policy outcome. A high score does not imply immunity.

Governance-heavy by design

The index is intentionally theory-led toward liberal-governance norms and is not weight-neutral. The governance dimension carries half the canonical score. Test the equal and governance-heavy weightings to see how much that assumption is doing.

Authoritarian comparability is imperfect

Information-environment controls in authoritarian systems may be state-driven rather than privately driven, which complicates cross-regime comparison of that dimension.

Edition sensitivity

WGI, V-Dem, and other providers revise their history. Scores can move between releases for a past year because a source was revised, not because conditions changed.

Revised economic dimension

The v2 economic dimension retires the legacy Doing Business formulation. It is not numerically identical to the original published chapter table.

Small featured set

This draft covers 16 countries. Percentiles, medians, and ranks describe that set only and will move as coverage widens.