General research for information only — not legal, tax, or investment advice. Cambodian law and markets change; figures are indicative, so verify current detail with a qualified local professional before you act.
The ASEAN+3 Macroeconomic Research Office (AMRO) released a fresh stress-testing exercise on Cambodia’s banking sector this week, and the headline finding sounds reassuring — the system can absorb non-performing loan ratios of up to 24% before capital adequacy falls below regulatory minimums. But the detail beneath that number carries specific, uneven consequences for anyone with property exposure in Cambodia.
This is the second AMRO assessment to land in as many days. Earlier in the week, AMRO warned that Cambodian banks’ dollar liquidity buffers at the National Bank of Cambodia had fallen to 10.4% of deposits — a record low — with roughly $10.5 billion shifted offshore. Together, the two reports frame a banking system that is capital-resilient in aggregate but structurally strained at the margin, and the property sector sits squarely in the strain zone.
What the Stress Test Actually Measures
AMRO ran a reverse stress test — a methodology that works backward from the regulatory minimum capital adequacy ratio (CAR) of 15% to determine how large a deterioration in asset quality the system could absorb before breaching that floor.
The headline result: Cambodia’s deposit-taking institutions could withstand an average NPL ratio of 24% before the system-wide CAR falls below 15%. Using a more conservative threshold that includes the capital conservation buffer — effectively a 17.5% CAR floor — banks could absorb NPLs of up to 18.6%, equivalent to roughly $6.3 billion in non-performing loans.
On first reading, those numbers are comfortable. The system-wide average NPL ratio stood at 8.3% at end-2025 (up from 7.1% in 2024 and 5.1% in 2023), meaning there is a substantial gap between current levels and the stress-test breaking point.
Where the Comfort Zone Ends
Three caveats in the AMRO analysis prevent the headline from reading as an all-clear signal, particularly for property investors.
1. The Aggregate Mask
A system-wide average capital adequacy ratio of 15% or above does not mean every bank meets that threshold. AMRO explicitly notes that the number of banks operating with limited buffers above the regulatory minimum increased between 2024 and 2025, with further pressure expected during 2026 as bad loans continue to rise.
Smaller deposit-taking institutions are disproportionately represented in this group. They generally operate with thinner capital buffers, weaker profitability, and higher existing NPL ratios — limiting their ability to absorb additional losses without raising fresh capital. For someone holding transaction deposits or a construction-finance facility with a smaller bank, the system-level resilience number offers limited comfort.
2. Sector Concentration
AMRO confirms that distressed loans remain concentrated in precisely the sectors that matter most to property investors: retail trade, agriculture, and real estate-related industries including construction, property development, and residential mortgages.
This is not a new finding — NBC’s H1 2026 banking assessment (published the same week) flagged construction and real estate for elevated NPLs, and the IMF’s July 2026 Article IV consultation warned that bank forbearance was masking the true extent of property-related credit stress. What the AMRO stress test adds is a quantification of the gap: the system could absorb roughly double the current NPL ratio, but only if the deterioration is evenly distributed. A concentrated shock in the property sector — say, a major developer default cascading through construction loans, supplier credit, and retail mortgages — would stress individual institutions well before the system-wide threshold.
3. The Liquidity-Complement Problem
The stress test measures capital adequacy — the ability to absorb losses on the asset side. It does not measure liquidity — the ability to meet withdrawal demands on the liability side. AMRO’s own companion report on dollar liquidity, published this week, shows that banks’ immediately accessible USD reserves at the NBC have fallen to their lowest level on record.
A bank can be solvent (capital adequate) but illiquid (unable to meet short-term obligations). In a dollarised system where the central bank cannot create US dollars, liquidity matters at least as much as capital adequacy. The stress test result does not change the liquidity picture, and the liquidity picture does not improve the stress test result — they are two independent constraints.
The Property-Specific Read-Through
For property buyers, developers, and investors in Cambodia, the AMRO stress test carries several practical implications.
Developer Financing
Developers relying on construction finance from smaller banks should be paying close attention to their lender’s capital position. If a bank’s CAR is near the regulatory minimum and its property-loan portfolio is substantial, any additional NPL deterioration could trigger a tightening of credit lines — precisely when a project under construction needs continued drawdowns.
The stress test result does not predict this outcome, but it establishes the conditions under which it becomes more likely. A developer whose bank is in the thin-buffer group should have a contingency financing plan.
Mortgage Availability
For individual buyers, the question is less about bank failure and more about credit availability. Banks with thin capital buffers and elevated property NPLs are likely to maintain a cautious approach to new mortgage lending, even if the system as a whole could absorb more losses. This means tighter loan-to-value ratios, longer approval timelines, and more selective underwriting — particularly for off-plan purchases, foreign buyers, and secondary-market resales where verification is harder.
Buyers who can finance through ACLEDA or one of the larger commercial banks will face fewer constraints than those relying on smaller institutions.
Distressed Asset Opportunity
For investors looking at the distressed-asset side of the market, the stress test is a useful calibration tool. The finding that the system could absorb $6.3 billion in NPLs before the conservative capital threshold is breached suggests that the distressed-asset pipeline — already flagged at approximately $1.5 billion in the MSC 2026 report — has room to grow without triggering systemic instability.
This does not make distressed assets safe (the underlying property and legal due diligence still matters), but it does reduce the systemic tail risk for buyers of NPL portfolios or receivership properties.
Putting the Stress Test in Context
The AMRO stress test should be read alongside the three other banking assessments published in July 2026:
| Assessment | Key Finding | Property Angle |
|---|---|---|
| AMRO Stress Test (Jul 29) | System can absorb 24% NPLs before capital breach | Concentrated in RE/construction; smaller banks exposed |
| AMRO Dollar Liquidity (Jul 28) | USD liquidity at record low, $10.5B offshore | Tighter credit availability for property lending |
| NBC H1 2026 (Jul 29) | 9.6% NPLs, 4.6% credit growth, $1.8B restructured | Construction/RE flagged as high-risk sectors |
| IMF Article IV (Jul 2026) | Bank forbearance masks credit stress | Warned of hidden NPLs and refinancing risk |
Together, these four assessments paint a banking system that is adequately capitalised but facing accumulating strain in its most property-exposed corners. The AMRO stress test is the most reassuring of the four — but only at the aggregate level, and only for capital adequacy, not liquidity.
What to Watch Next
Three indicators will determine whether the stress test’s margin of safety narrows or widens in the coming months:
-
NPL trajectory for H1 2026 — NBC’s official H1 NPL data (which it reported at 9.6%) is the real-time gauge. If H2 2026 shows further deterioration toward 11-12%, the distance to the 18.6% conservative threshold narrows significantly.
-
Small-bank capital raises — The number of smaller institutions raising fresh capital (or merging) over the next 6-12 months will signal whether the thin-buffer group is self-correcting or deteriorating.
-
Developer defaults — A single large developer default in Phnom Penh or Sihanoukville would be the most informative stress event, revealing whether the system-level resilience holds at the individual-institution level.
For property investors in Cambodia, the AMRO stress test is neither an all-clear nor a warning — it is a map of where the fault lines run. The system is resilient, but the resilience is distributed unevenly, and property sits on the more vulnerable side of that distribution. Due diligence on your counterparty bank and developer has never been more relevant.
This article is part of Research Cambodia’s ongoing coverage of banking-sector risk and its implications for property investment. See also the companion analysis on AMRO’s dollar liquidity warning, the NBC H1 2026 banking assessment, and the IMF Article IV forbearance analysis.
Frequently asked questions
What did the AMRO stress test actually find about Cambodia banks?
The ASEAN+3 Macroeconomic Research Office reverse stress test found that Cambodia deposit-taking institutions could withstand NPL ratios of up to 24% before average capital adequacy falls below the NBC minimum of 15%. Using a more conservative benchmark including the capital conservation buffer, the threshold is 18.6% NPLs — equivalent to roughly $6.3 billion in bad loans.
Is the stress test result good or bad news for property investors?
Both. System-level resilience is reassuring — no imminent systemic banking crisis. But the stress test also confirms that real estate, construction, and residential mortgages remain the largest source of non-performing loans, and smaller banks with concentrated property exposure face thinner buffers. The sector is resilient at aggregate level, not at every individual institution.
How does the AMRO stress test compare with NBC data?
NBC reported 9.6% NPLs in its own H1 2026 assessment, with construction and real estate flagged as high-risk. AMRO puts the system-wide average at 8.3% for year-end 2025. Both numbers confirm rising bad loans. The stress test is complementary — it answers the question of how much worse it could get before triggering capital adequacy breaches.
Which banks are most vulnerable to property-related credit risk?
AMRO notes that smaller deposit-taking institutions with thinner capital buffers, weaker profitability, and higher existing NPL ratios are significantly more exposed. The number of banks operating with limited buffers above the regulatory minimum increased between 2024 and 2025. Distressed loans remain concentrated in retail trade, agriculture, and real estate-related industries including construction, property development, and mortgages.
Does this mean I should avoid Cambodian property?
Not necessarily — but it reinforces the case for due diligence on your chosen bank and developer. The stress test is not a warning of imminent failure but a measure of systemic resilience. For buyers, it means checking whether your lender is among the smaller institutions with thin buffers. For developer deals, ensure construction financing is coming from well-capitalised banks. Lender health is now a legitimate part of property due diligence in Cambodia.