# Appendix N — Research Methodology and Verification Standard This appendix records how the report was researched and checked, and — as importantly — what could not be established. A reader who intends to rely on any statement in this report should read this appendix first, because it defines the weight each class of statement is entitled to carry. ## N.1 The standard applied Every legal, regulatory and fiscal claim in this report is either cited to a primary or official source, or expressly identified as requiring confirmation at the date of application. There is no third category. Where a proposition could not be established from an authoritative text, the report says so rather than asserting it in softened language. Two consequences follow, and both are deliberate. The report is longer than a marketing document, because stating a limitation takes more words than omitting it. And the report declines to answer some questions a prospective applicant would like answered — most visibly the practical travel access conferred by the São Tomé and Príncipe passport, which is discussed in §6.9 in the only terms the available sources support. ## N.2 Source hierarchy Sources were ranked, and where they conflicted the higher rank governed: 1. Primary legislation, official gazettes and court judgments — the decree-law and nationality statute of São Tomé and Príncipe as published in the Diário da República; Greek legislation as published in the Government Gazette (Φύλλο Εφημερίδας της Κυβερνήσεως); judgments of the Court of Justice of the European Union as published by the Court. 2. Official government portals and administrative guidance — the Greek Ministry of Migration and Asylum, the Greek independent tax authority, the São Tomé Citizenship Investment Unit, and the equivalent authorities of the comparator jurisdictions. 3. The publications of the European institutions, the Financial Action Task Force, the Organisation for Economic Co-operation and Development, the Hague Conference on Private International Law, central banks and national statistical offices. 4. Reputable secondary sources — professional firms, established financial and legal press — used for market context and practice only, and never as the sole authority for a legal or fiscal proposition. Commercial marketing material, intermediary websites and passport-ranking indices were not treated as sources. Where such material is the only origin of a figure in circulation, the report either omits the figure or identifies it as unverified. Instruments were read in their original languages. The São Tomé instruments were read in Portuguese from gazette facsimiles; Greek legislation and administrative guidance were read in Greek, in consolidated form where a consolidation exists. ## N.3 The verification process The research was conducted in four passes, each performed independently of the last: **Pass one — primary-source research.** Ten subject domains were researched separately: São Tomé nationality and investment law; Greek investor-residence law; the end of European investor citizenship; Schengen mobility law; the financial-crime and tax-transparency literature; Greek taxation; the Greek residential property market; banking, common-reporting-standard mechanics and succession; the nationality law and exchange-control position of the client jurisdictions; and Kestrel Private's own published material. Each domain produced a table of numbered core claims, each claim tied to the instrument, article and access route relied on. **Pass two — adversarial verification.** Every domain was then re-checked in a second, separate verification pass whose instruction was to refute rather than confirm: to open each cited source independently, to prefer a primary text wherever a secondary source had been relied on, and to record any claim whose source could not be opened as unverifiable rather than accepted. Approximately 370 core claims were checked in this pass. Corrections were recorded with the corrected text and its source; no claim was upgraded from reported to verified without the underlying instrument being read in that pass. **Pass three — consolidation.** The verified material was consolidated into a single fact base of individually referenced propositions, separated into confirmed facts and a distinct register of items that verification had left unestablished. Claims that verification refuted were removed entirely rather than softened. Where researcher and verifier disagreed and the disagreement could not be resolved from the text, both positions were recorded in the unestablished register. **Pass four — drafting, challenge and revision.** Chapters were drafted only from the consolidated fact base, so that no chapter could introduce a fact that had not survived verification. Each chapter was then read by a reviewer instructed to attack it from four directions: as opposing counsel searching for an actionable overclaim or implied guarantee; as a regulator reading the document as a financial promotion; as a rival researcher hunting for a factual error; and as a sceptical client's lawyer testing whether each citation genuinely supports the sentence it is attached to, or merely sits near it. Every sustained finding was applied. Findings that were rejected were recorded, with reasons, rather than discarded. ## N.4 Classification of statements - **Statements of law** are made in the plain indicative and carry a citation to the instrument and article. They are accurate as at 2 August 2026 and no later. - **Statements of administrative practice** are identified as such. Practice is more volatile than law and is recorded with the date of the administrative record consulted. - **Statements of market fact** — prices, yields, indices, transaction costs — carry their statistical source. Where a figure represents a market norm rather than a published statistic, it is given as a band and identified as reported rather than verified. - **Illustrations and scenarios** are labelled. Every assumption behind a scenario is stated alongside it. No scenario is a forecast, and the report does not claim predictive value for any of them. - **Items requiring confirmation** are those the verification process could not establish. They appear with that qualification attached, in the same sentence, every time they appear. ## N.5 What could not be established The following limitations are material and are repeated here so that they are not buried in the chapters: - **Greek taxation.** The Greek tax positions in this report rest on the tax authority's published guidance and on professional summaries of the governing statutes rather than, in every case, on the gazette text of each amending law. The rates and mechanics stated are those in force according to those sources as at 2 August 2026; they are stated to the standard of reliable guidance rather than to the standard of a read gazette, and they require confirmation from a Greek tax adviser before any client relies on them. - **The São Tomé programme's official portals.** For most of the research period neither of the programme's published web presences was reachable directly, and both were consulted through archived captures: cip.gov.st at captures of 10, 17 and 28 July 2026, and stpcip.com at a capture of 19 April 2026. The government domain cip.gov.st was subsequently reached live and read on 4 August 2026, and the former domain stpcip.com now redirects to it. The post-approval document charge of US$750 used in the cost model, and the contribution and submission-fee schedule alongside it, therefore rest on that live official-domain read rather than on the April 2026 capture (§10.4). Statements drawn from captures rather than from the live read are accurate as at the capture cited in the relevant note, not as at the date of this report; every programme charge is identified as requiring confirmation at the date of application wherever it appears. This limitation is one reason the report insists throughout that the programme's operational status be confirmed at the date of application. - **Published fee schedules of the citizenship programme.** The gazetted fee annex does not itemise post-approval document charges. The report therefore uses the official aggregate figure and identifies the more detailed breakdowns in circulation as unverified. - **Divergences between the gazette and the programme's published material.** In at least one respect — the definition of qualifying dependants — the administrator's published material is wider than any provision the research could locate in the decree-law. The report records the divergence instead of resolving it, because only the responsible authority can resolve it. - **Travel access.** No authoritative, citable enumeration of the visa-free access conferred by the São Tomé passport was found. The report states only what could be verified against destination-state sources. - **Practice at the Schengen external border.** The interaction between a residence permit issued against one passport and a subsequently acquired second passport is documented in the published guidance of some member states but is not the subject of a single uniform published rule. The report describes it as practice, identifies the guidance relied on, and recommends the conservative course. - **Market and cost bands.** Professional-fee ranges, renovation costs, insurance premiums and time-to-sell estimates are market observations, not published statistics. They are given as bands, identified as reported, and should be replaced by written quotations before any client commits. ## N.6 Currency of the report Law, fees, administrative requirements and market data change. Three dates govern this report and they do different work. The **research cut-off is 2 August 2026**: every statement of law, every official source and all market data are made as at that date, and it is the date to cite when asking what this report establishes. A small number of **fee and administrative-practice confirmations were re-checked to 5 August 2026** and carry their own date where they appear, so that a later confirmation is never mistaken for a wider re-verification. The **publication date is 6 August 2026**. Where those dates differ, the research cut-off governs. The report is not maintained as a live document, and no reader should treat it as evidence of the position at any later date. The programmes it describes are both subject to change by the responsible governments, in one case expressly by ministerial order; see §14.1 and §14.2. ## N.7 Corrections Kestrel Private will correct any error of law, citation or figure that is drawn to its attention and can be established against a primary source. A report that argues for the primacy of primary sources cannot reasonably decline to be held to them. The verification method and standard set out above are what this appendix publishes. The pass-by-pass record of findings raised, applied and rejected during review is retained as an internal assurance record and is available to a publisher, regulator or professional adviser on request.