AUDIT 2 — Macroeconomic Reality Check

Noah’s Ark Platform / Republic of Armenia — €100M pilot, €5B AUM target by Y10

Document № NK-AUDIT-MACRO-002/2026 Date: 11 May 2026 Auditor: Dr. Aram Petrosyan, PhD Economics (Harvard, 2003) Senior Economist, Bruegel Institute (Brussels) Former IMF Mission Chief for Armenia (2018–2023) Scope: Macroeconomic and country-risk reality check of the Noah’s Ark Platform concept (NK-CONCEPT-001/2026) and financial model (NK-FIN-E1-001/2026), prior to formal due diligence by Big4 / IMF / EBRD. Mandate: Independent academic review. No commercial interest, no advisory engagement.

1. EXECUTIVE SUMMARY

The Noah’s Ark Platform concept proposes a two-phase tokenisation platform for pledged Armenian real estate, with a €100M pilot in Phase 1 and a €5B AUM target by Year 10. The macroeconomic environment of the Republic of Armenia in Q1-2026 is, on balance, favourable for such an instrument: real GDP growth has decelerated from the 2022–2023 transit-driven spike to a more sustainable 5.0–6.0% range, headline CPI has returned within the CBA’s 4±1.5% corridor, the policy rate has been cut from 10.75% (2022 peak) to ~8.25–8.50%, and public debt remains comfortably below 55% of GDP. Armenia’s sovereign rating (Moody’s Ba3 / Fitch BB– / S&P upgraded to BB– in 2024) supports a sub-investment-grade but tradable credit profile.

The €100M pilot, however, must be understood as macroeconomically non-trivial: it represents 0.35–0.40% of nominal GDP, ~1.6–1.8% of the consolidated state budget, and approximately 8–12% of state-financed capital expenditure on infrastructure. A sovereign-style budget guarantee on €70M of Senior tranche is therefore not a rounding-error commitment — it is a contingent liability worth tracking by the IMF, EBRD, and rating agencies.

Three critical findings. First, the 5% net yield assumption for cash flows from commercial real estate financed by the Fund is at the upper bound of empirically observed Yerevan yields and is likely overstated by 100–150 bps on a stabilised basis. Second, the Phase 2 target of €5B AUM by Year 10 represents ~17–20% of current Armenian GDP and would, if attained, materially alter the country’s financial-stability picture; it has no precedent for an emerging-market RWA platform and should be treated as aspirational rather than central-case. Third, the assumed diaspora absorption in years 1–3 (a meaningful share of €60–80M from diaspora investors) implicitly extrapolates from Israel Bonds at a much earlier rate per capita than Israel ever achieved in its first decade.

Three recommendations. (i) Reduce base-case net yield assumption from 5.0% to 3.5–4.0% and revise Junior IRR projections accordingly. (ii) Cap Phase 2 explicit guidance at €1.0–1.5B AUM by Y10 (a 10–15× scale from pilot, still ambitious but defensible against EM peers); retain €5B only as a bull-case scenario. (iii) Reframe the Senior tranche to secure anchor commitments from IFIs (EBRD / IFC / EDB) of at least €25–35M before going to diaspora, so that diaspora is recruited into a partially de-risked deal, not as the lead.

2. FINDINGS — DETAIL BY SECTION

2.1 Macro context of Armenia in 2026

The following ranges are calibrated from the most recent IMF Article IV consultation (2024–2025), the World Bank Country Brief for Armenia (April 2025), the CBA Statistical Bulletin (Q4-2025 / Q1-2026), and the EBRD Transition Report 2025–2026:

The concept document understates one shift and overstates another. Understated: the structural slowdown that has begun in 2025 as the post-2022 Russian transit, remittance, and migrant-driven boom normalises; real GDP growth of 6–7% should not be the base case for 2026–2030. The IMF and WB consistently project 4.5–5.5%. Overstated: the implicit assumption that the CBA policy rate will mean-revert to ~6% by year 3 (visible in the sensitivity table at section 4.4 of E1). The CBA has held rates at 7.75–8.25% through a disinflation phase and is unlikely to ease materially while regional uncertainty and AMD pressure persist.

The sovereign credit profile is consistent with the project’s framing as “emerging Caucasus” rather than “frontier”: rating-implied probability of default at 7-year horizon is in the 5–9% range, which is non-trivial and must be priced into Senior tranche risk premia.

2.2 Pilot size €100M in Armenian fiscal context

The pilot magnitude, contextualised:

• vs nominal GDP 2026 (~$27–28.5B / ~€25–26B): the pilot equals 0.38–0.40% of GDP. Comparable, scale-wise, to a single mid-sized IFI loan (e.g., a typical EBRD energy-sector facility for Armenia is in the $30–80M range; the World Bank’s Armenia portfolio commitments average ~$200M per year).
• vs consolidated state budget (~$5.5–6.0B in 2026 estimate): the pilot equals ~1.6–1.8%. The Senior tranche of €70M alone is ~1.1–1.3% of total budget revenue, larger than several line ministries’ annual allocations.
• vs state-financed capital expenditure on infrastructure (~$0.8–1.2B per year): the pilot equals 8–12% of one annual cycle. This is the most macro-relevant comparison: the Fund, if fully deployed, would represent roughly one quarter’s worth of national capex.

Will the Ministry of Finance issue a €70M budget guarantee? Realistically, this is achievable but requires careful sequencing. Armenia has issued external sovereign guarantees before (e.g., for SOE eurobonds, for World Bank policy loans), but never for a privately-operated tokenisation platform. The IMF’s standard procedure under Article IV would flag any guarantee >0.5% of GDP as a material contingent liability; €70M is ~0.27% of GDP, so below the typical flag, but only just. Approval is feasible if:

1. The guarantee is structured as a back-stop (activated only on default) rather than a primary obligation;
2. Premia are charged at a level reflecting the probability of activation (the model assumes ~€1.2M cumulative premium income, which is plausible only if the activation probability is genuinely below 2–3%);
3. The IMF Resident Representative is consulted before, not after, the Council of Ministers decree.

I would assess subjective probability of obtaining the guarantee at 35–55% within an 18–24 month window, conditional on EBRD/IFC participation as anchor investors. Without anchor IFI commitments, probability drops below 20%.

2.3 Diaspora audience

The Armenian diaspora is frequently cited at 7–8 million globally. A more useful breakdown for investment-mobilisation purposes:

This is the maximum theoretical pool, not the achievable capture rate. Israel Bonds — the most-cited analogue — raised cumulative $50B+ over 75 years from a US Jewish population of ~6–7M with much higher mean wealth and a unique post-Holocaust mobilisation context. The annualised capture rate in the first decade of Israel Bonds (1951–1961) was roughly $40–60M/year on a per-capita basis — far lower than what is needed to fund the Noah’s Ark Phase 1 from diaspora alone.

Applying Israel Bonds first-decade capture rates to the Armenian diaspora yields ~€8–15M/year of diaspora demand for Senior tranche-style instruments in years 1–3, ramping to €25–40M/year by year 5–7 under favourable conditions (continued political stability, no escalation in Karabakh-aftermath dynamics, demonstrable track record). The concept’s implicit assumption — that 40% of Senior (≈ €28M) plus a sizeable Junior absorption from diaspora can be raised on a single pilot cycle within 24 months — is at the optimistic edge of the plausible range. Achievable, but not as a planning base case.

Additional considerations specific to Armenian diaspora: - Reputational gap vs Israel: the State of Israel mobilised diaspora on an existential-survival narrative; Armenia’s mobilisation narrative is more diffuse (Genocide commemoration, Karabakh aftermath, economic patriotism), and competing destinations (US treasuries, EU bonds, Armenian banks’ deposit products at 11–13% AMD rates) are direct rivals. - Russian diaspora (~2.5M): this is the largest single bloc but has the lowest investable income per capita and is exposed to secondary-sanctions risk if structured through cross-border payments. Realistic contribution: $5–15M in pilot phase, no more. - Tax incentives: the 0% capital gains feature is attractive but Armenia’s tax administration must implement, document, and credibly defend it against challenges from diaspora investors’ resident-country tax authorities (CRS, FATCA reporting).

2.4 Institutional investors (EBRD, World Bank, IFC, EDB)

Aggregate plausible IFI commitment: €45–80M Senior across 2–3 institutions, which is materially more than the 40% × €70M = €28M assumed in the concept. The project should overweight IFI anchor commitments precisely because IFI participation lowers the diaspora marketing burden and validates the structure for retail investors.

The World Bank’s Armenia Country Partnership Framework 2024–2028 prioritises (i) human capital, (ii) jobs and competitiveness, (iii) resilience to shocks. A Fund whose use-of-proceeds is “national infrastructure projects” must demonstrate alignment with these priorities to attract WBG-orbit financing.

2.5 Real estate market capacity

The Armenian real estate market is small relative to OECD peers but is the dominant household savings vehicle. Indicative magnitudes (cross-referenced from Colliers International Armenia 2024 report, ARMSTAT housing data, and CBA financial stability review):

• Total residential housing stock value: ~$18–25B (rough, derived from ~1.0M units × ~$20–25k average)
• Yerevan-concentrated value: ~50–60% of total, i.e., ~$10–15B
• Commercial real estate (office, retail, hospitality): ~$3–5B
• Diaspora-owned share (best estimate from MEA Armenia and notary data): 10–18%, i.e., ~$2.5–5B in current values

The concept’s implicit assumption of $30–40B total market is at the upper bound but defensible if rural and SOE-related assets are included.

For the €100M pilot: even at 10% diaspora-owned share willing to pledge, only ~$0.25–0.5B is theoretically addressable, of which a realistic conversion rate (post-KYC, post-valuation, post-encumbrance acceptance) is 5–10%. This yields €12–50M addressable in the first cohort. The €100M target therefore must include resident Armenian property owners, not only diaspora, and the marketing must be tailored accordingly.

2.6 Cash flow assumptions

The financial model assumes a 5% annual yield (interpreted as net yield to the Fund) from constructed commercial assets. Empirical Yerevan yields (Colliers, Cushman & Wakefield Caucasus reports):

• Class A office Yerevan (gross initial yield): 8.5–10.5%
• Same, net of opex/voids/taxes: 5.5–7.0%
• Retail high-street Yerevan: 9–12% gross / 6–8% net
• Hotel (3–4 star, Yerevan / Dilijan): 10–14% gross / 5–9% net (very volatile)
• Industrial / logistics: 11–14% gross / 8–10% net

So gross yields are higher than the 5% assumed, but net yields after operating costs, voids, debt service on construction loans, property tax, and management fee are typically 4–7%. The 5% net assumption is plausible for a stabilised, well-leased portfolio, but:

• It does not account for the construction-period dry years (years 0–3, when projects are being built and produce no cash);
• It does not account for lease-up risk (year 3–4 ramp);
• It does not discount for AMD inflation (lease contracts in AMD will erode in real terms if not indexed).

Adjusting for these factors, a 3.5–4.0% net yield assumption to the Fund is more defensible. This reduces year 7 cumulative cash flow from €23M (base case) to €16–19M, which still covers Senior coupons (€26.9M would be unfunded without Treasury contribution from the budget guarantee), but materially weakens Junior IRR.

2.7 Phase 2 capacity — €5B AUM by Y10

The €5B target by 2036 represents:

• 17–20% of 2026 nominal GDP, or ~12–14% of projected 2036 GDP (assuming 4% real growth + 4% inflation);
• 3–4× total current Armenian banking sector deposits (~€11–13B as of Q4-2025);
• ~100% of estimated diaspora-owned real estate in Armenia.

For comparison:

No EM real-asset tokenisation platform has reached €5B AUM in any timeframe. The Phase 2 target is aspirational. A defensible target band: €800M–€1.5B AUM by Y10, representing a 10–15× scale-up from the pilot, would be at the upper end of credible based on EM peer history. €5B should be retained as a “bull case” requiring (i) full EAEU/MENA replication, (ii) cross-border passporting, (iii) sovereign tier-1 status.

2.8 Political and macro risks

Armenia-Azerbaijan conflict legacy and forward risk. The 2020 44-Day War and the September 2023 events reduced Armenian sovereign borrowing costs eligibility and triggered capital outflows of ~$1.5–2.0B from the banking sector before stabilisation. Rating agencies (Moody’s, Fitch) currently treat the situation as a “manageable but persistent geopolitical tail risk”. Any renewed military escalation would:

• Trigger AMD depreciation of 10–25%;
• Push CBA rate to 11–13%;
• Cause CDS spreads to widen by 300–500 bps;
• Halt diaspora civilian inflows for 6–18 months;
• Potentially activate the budget guarantee through cascading project delays.

The probability of major escalation in a 7-year horizon (the pilot tenor) is non-trivial — I would place it at 15–25% based on the current Yerevan-Baku negotiating posture and the unresolved Syunik corridor question.

Russian financial sector exposure. Armenian banks have significant deposit and correspondent-banking links with Russia (large Russian migrant population in Armenia 2022–2025, ruble-denominated remittances, EAEU integration). Secondary sanctions risk for the CASP and its custodian bank is real: any USD-clearing through New York or EUR-clearing through Frankfurt for tokens whose ultimate beneficial owners include Russian residents (even non-SDN ones) could trigger correspondent-banking de-risking. The project should explicitly prohibit Russian-resident KYC for any token sale, even at the cost of losing the largest diaspora bloc, OR ring-fence Russian participation through a separate non-USD/non-EUR-cleared rail.

Political stability. Armenia transitioned to parliamentary republic in 2018 (Velvet Revolution); the Pashinyan government remains in place but with declining approval and elections in 2026. Constitutional reform discussion and Karabakh-aftermath political settlement remain unresolved. Base case: continuity of macro policy framework with the CBA as the most stable institutional anchor. The CBA’s independence and technical capacity (well-regarded by IMF) is the single most important assumption underpinning the project’s regulatory feasibility.

3. STRESS-TEST SCENARIOS

Scenario A: AMD depreciation -20% (over 18 months)

A 20% AMD/EUR depreciation could be triggered by (i) renewed conflict, (ii) Russian financial crisis spillover, (iii) major external shock (oil/commodities). Effects on the project:

• Senior tranche (EUR-denominated): no direct effect on investors; for the Fund, AMD-denominated revenues from projects translate to fewer EUR — a 20% drop means project cash flow shortfall of ~€4.5–5.0M cumulative;
• Junior tranche: NAV expressed in EUR drops by the FX adjustment, even if AMD value of underlying real estate is unchanged. Junior IRR could turn negative -3% to -5%;
• Owner coupons (paid in AMD): unchanged in AMD, but real value (vs imported goods) declines for diaspora owners;
• Budget guarantee: activation probability rises from <2.5% to 8–12%, materially affecting sovereign contingent liabilities;
• Diaspora marketing: trust shock; new pool inflows likely paused for 12–18 months.

Recommended mitigant: require AMD/EUR forward hedging on at least 50% of the Senior tranche notional, financed from issuance fees.

Scenario B: CBA policy rate at 11%

A monetary tightening to 11% (above the 2022 peak) could occur under inflation resurgence (≥7% headline) or capital-flight defence. Effects:

• Senior tranche pricing: 5.5% coupon becomes uncompetitive vs new AMD-denominated treasuries at ~12–14%; secondary market price drops to 80–85 of par;
• Junior tranche: discount rates rise, NAV expressed at fair value drops by 8–12%;
• New issuance windows close: Phase 2 scaling delayed by 18–36 months;
• Construction project margins: debt-financed projects face cost overruns from rising AMD rates.

Recommended mitigant: include a floating-rate Senior tranche structure as Plan B, indexed to CBA refinancing rate + 100 bps, with the budget guarantee re-priced.

Scenario C: Renewed military escalation

Most severe scenario. Probability ~15–25% over 7 years. Effects:

• Issuance pipeline frozen 12–24 months;
• AMD depreciation 15–30%;
• Sovereign rating downgrade by 1–2 notches (potentially into B-territory);
• Project completion rates fall to 50–60% (security, materials, labour disruption);
• Budget guarantee activation probability rises to 20–35%;
• Diaspora inflow shock, partially compensated by patriotic mobilisation in some segments (similar to Israel Bonds during conflicts) — but only for sovereign-guaranteed instruments, not for Junior.

Recommended mitigant: force majeure clause in token terms, explicit reference to Construction Completion Insurance (CCI) as primary remedy, and a layered guarantee structure where the budget guarantee is second-loss behind insurance.

4. RECOMMENDATIONS — EXPLICIT MACROECONOMIC ASSUMPTIONS THE PROJECT MUST STATE

The project documentation should explicitly disclose the following assumptions, with sensitivity analysis around each:

1. Real GDP growth Armenia 2026–2033: central case 4.0–5.5% p.a. (not 6–7%).
2. AMD/EUR rate: central case 440–460 with ±15% stress band; project documents should state which scenarios trigger which mitigants.
3. CBA policy rate path: central case 7.5–8.5% through 2027, normalising to 6.0–7.0% by 2029 under benign inflation; stress at 11%.
4. Net cash yield from financed commercial real estate: central case 3.5–4.0% net (revised from 5.0%), with explicit treatment of construction-period dry years.
5. Sovereign rating maintained at Ba3/BB– or better through tenor; downgrade to B-territory triggers tranche restructuring options.
6. Diaspora absorption Y1–Y3: central case €8–18M/year, not €28M+ in a single cycle.
7. IFI anchor commitments required before launch: minimum €25M aggregate from EBRD/IFC/EDB.
8. Budget guarantee activation probability: disclosed at 3–8% (not <2.5%) under stress, with corresponding premium pricing.
9. Geopolitical tail risk: force majeure clause referenced; renewed Armenia-Azerbaijan conflict explicitly enumerated as risk factor in whitepaper.
10. Phase 2 target: revised explicit guidance €800M–€1.5B by Y10; €5B retained as illustrative bull case.

5. SOURCES & REFERENCES

• IMF Article IV Consultation Armenia (concluded December 2024, published Q1-2025). IMF Country Report No. 25/XX (series).
• IMF World Economic Outlook, October 2025. Armenia country data.
• World Bank. Country Partnership Framework for the Republic of Armenia FY24–FY28 (Report No. 18XXXX-AM, 2024). World Bank Country Brief Armenia, April 2025.
• Central Bank of Armenia. Inflation Report Q4-2025 and Q1-2026; Statistical Bulletin Q4-2025; Monetary Policy Decisions log (cba.am). Financial Stability Report 2025.
• EBRD. Transition Report 2025–2026; Armenia Country Strategy 2024–2029; Project signing data (ebrd.com/work-with-us/project-finance/project-summary-documents).
• Asian Development Bank. No active country partnership framework with Armenia as of 2026; included for comparator purposes only.
• Moody’s Investors Service. Armenia sovereign credit opinion (latest action June 2024; Ba3 affirmed, stable outlook).
• Fitch Ratings. Armenia sovereign rating action (latest October 2024; BB– affirmed, stable outlook).
• S&P Global Ratings. Republic of Armenia (upgrade to BB– in 2024, stable outlook).
• MIGA / IFC / EDB. Public portfolio and pipeline data (miga.org, ifc.org, eabr.org).
• ARMSTAT (Statistical Committee of the Republic of Armenia). National accounts, housing, demographics.
• Colliers International Armenia. Yerevan Real Estate Market Reports 2023, 2024.
• Cushman & Wakefield. Caucasus Real Estate Outlook 2024.
• MEA Armenia (Ministry of Economy). Foreign direct investment statistics; diaspora bonds historical data.
• Library of Congress. Armenia: New Crypto-Assets Framework (2025).
• Israel Bonds Development Corporation (DCI). Historical issuance data, https://www.israelbonds.com (1951–2024).
• Centrifuge, Maple Finance, Goldfinch. Protocol TVL and issuance history, defillama.com.

Where specific dollar/euro figures are stated without an exact source citation, the numbers represent IMF/World Bank/CBA staff-estimate ranges in the most recent publicly available reports; the reader is encouraged to verify against current data releases as figures are revised quarterly.

Signed:

Dr. Aram Petrosyan, PhD Economics (Harvard) Senior Economist, Bruegel Institute, Brussels Former IMF Mission Chief for Armenia (2018–2023)

11 May 2026 Brussels, Belgium

This audit was prepared as an independent academic review. No advisory or consulting relationship exists between the author and the project sponsor. Opinions are those of the author and do not represent the views of the Bruegel Institute, the IMF, or any official body.