The Foreign Currency Mortgage Trap in 2026: Lessons from the Swiss Franc Disaster, the Yen Carry Risk, and Why Hard-Currency Debt Is Quietly Returning
Published on: May 12, 2026
Quick answer: A foreign-currency mortgage bundles a low-rate loan with an implicit short position on the funding currency against the currency you earn in, and the exchange rate, not the headline coupon, decides whether you keep the property. When the franc revalued after the SNB removed its EUR/CHF floor in January 2015, Central European households who had borrowed in CHF at 2–3 percent saw their balances surge in local-currency terms on properties that had not appreciated, triggering years of litigation. The same trade is now being remarketed as a roughly 1.5 percent yen mortgage. The decisive test is whether you have natural income in the loan currency: if rental or earned income in that currency services the debt, it is a hedge; if not, it is leverage on an FX bet, so stress-test the loan principal in your earning currency after 20 and 40 percent adverse moves before signing.
TL;DR, In 2008, roughly 1.7 million Hungarian, Polish, Croatian, and Romanian families took out Swiss franc mortgages at headline rates of 2–3 percent. By 2015, after the SNB removed its EUR-CHF floor, many of them owed 60–80 percent more in local-currency terms on a property that had not appreciated. Two decades later the same trade is being marketed to international property buyers in a new wrapper: the Japanese yen mortgage at 1.5 percent. The 2026 framework for cross-border buyers thinking about borrowing in a currency they do not earn, when it works, when it kills you, and why the "natural hedge" is the only version of this trade that survives a full currency cycle.
The same trap, rebuilt in a new currency
Every cross-border real estate cycle produces some version of the same trade. A buyer in a high-interest-rate jurisdiction is offered a mortgage denominated in a low-interest-rate currency. The headline coupon looks irresistible. The buyer signs. For a few years it works. Then the funding currency strengthens, the buyer's local currency weakens, and the mortgage balance, converted back into the buyer's earning currency, quietly doubles.
This is the foreign-currency mortgage trap. It is one of the most predictable wealth-destruction patterns in international real estate. It has happened, in slightly different forms, in Hungary, Poland, Croatia, Romania, Iceland, Turkey, and parts of Latin America. In 2024–2026, the marketing pitch is being reassembled around the Japanese yen, and a new generation of cross-border property buyers, who have never lived through a currency crisis, is being offered the same product their parents took out twenty years ago.
This article is the JanusHermes 2026 framework for foreign-currency mortgages: how the trap actually works mathematically, the historical case studies that still matter, the 2026 yen carry context, and the narrow set of circumstances in which a foreign-currency mortgage is the correct answer rather than a postponed disaster.
The mechanics of the trap
A foreign-currency mortgage is, in its simplest form, two trades bundled into one product:
- A fixed-or-floating-rate loan denominated in a foreign currency.
- An implicit short position on that foreign currency against your earning currency, with notional equal to the loan balance.
The first trade is what the borrower thinks they are doing. The second trade is what determines whether they keep their property.
The arithmetic is unforgiving. Suppose a Hungarian household earns forints (HUF), and in 2007 took out a HUF 30 million-equivalent Swiss franc mortgage at 3 percent, against a property worth HUF 30 million. The CHF/HUF exchange rate at signing was approximately 150. So the loan principal in CHF terms was approximately 200,000.
By 2015, after the Swiss National Bank removed the EUR/CHF floor in January and the franc revalued sharply, CHF/HUF rose to roughly 300. The 200,000 CHF principal had not changed. But converted back to forints, which is what the borrower had to do every month to make the payment, it now represented HUF 60 million on a property still worth HUF 30 million.
The household was, in everyday terms, underwater by 100 percent of the original property value before a single late payment. The interest coupon of 3 percent had never been the relevant number. The exchange rate was.
This is what happened across Central Europe in 2008–2015, and what happened on a smaller scale in Croatia, in Romania, and in Poland. Estimates of the total number of households affected vary, but the consensus range is 1.5 to 2 million CHF mortgage borrowers across the region, of whom several hundred thousand defaulted or required government-mandated restructuring.
The post-mortem produced years of litigation, regulatory rewrites, and, in Poland, a multi-year court-driven mass restructuring of the so-called frankowicze, the "franc-borrowers", that is still working through the system in 2025–2026.
The historical case studies that still matter
Hungary
Hungary is the textbook case. By 2008, an estimated two-thirds of all retail mortgage debt in Hungary was denominated in Swiss francs. The exposure ran to roughly 5 percent of Hungarian GDP. When the franc strengthened, the household debt service ratio rose dramatically, and the government was eventually forced to step in with a 2014–2015 mass forint-conversion program that crystallized the losses on bank balance sheets.
The Hungarian outcome: the banking system was recapitalized, the borrowers received partial relief, and the country imposed a permanent cap on foreign-currency lending to households earning in forints. Hungary, in regulatory terms, learned the lesson. The cost was approximately a decade of suppressed homeownership and a generation of household financial trauma.
Poland
Poland's experience was structurally similar but politically louder. Polish frankowicze (an estimated 700,000 borrowers at peak) became a national political constituency, and the legal system has, since 2019, been progressively striking down CHF mortgage contracts as containing abusive clauses. Polish banks have collectively taken provisions running into the tens of billions of zlotys against this litigation, and the workout continues in 2025–2026.
The Polish lesson: even when the borrower legally won, the recovery took fifteen years. The household that signed a CHF mortgage in 2008 and prevailed in court in 2023 had spent the intervening period in litigation, with the property typically frozen and credit damaged.
Croatia and Romania
Croatia's kuna-CHF episode and Romania's leu-CHF and leu-EUR episodes followed broadly the same pattern. Croatia eventually pushed through a 2015 conversion of CHF mortgages to euros at a politically negotiated rate. Romania's CHF problem was smaller but the euro-mortgage problem, for households earning in lei, is structurally identical and has played out in slow motion.
The common thread across all four cases: the marketing pitch was "lower interest rate." The underlying product was a leveraged short position on the local currency.
The 2026 yen carry parallel
Through 2024 and into 2026, the Japanese yen has been the developed world's lowest-yielding major funding currency. Bank of Japan policy normalization has begun but has proceeded slowly. Yen-denominated mortgages, at headline rates between approximately 1.0 and 2.0 percent, have been actively marketed by certain Japanese and Hong Kong banks to international property buyers, particularly to buyers in Hong Kong, Singapore, Taiwan, and parts of Southeast Asia looking at Japanese property under the akiya wave and the broader 2024–2026 Tokyo and Osaka boom.
The pitch is the same as it was in 2007: "Why pay 5 percent in your home currency when you can pay 1.5 percent in yen?"
The mechanics are identical to the CHF trade. A Singaporean buyer who finances a JPY 100 million Tokyo property with a yen mortgage is implicitly short JPY/SGD with notional equal to the mortgage balance. If the yen strengthens 20 percent against the Singapore dollar over the loan term, which is precisely what happened to the franc against the forint, and which yen-dollar history suggests is well within the cycle, the borrower will owe 20 percent more in Singapore dollar terms on a property whose Tokyo yen price has not changed.
The borrower in this case will probably survive. JPY 100 million is, for a typical Singaporean priority-banking client, a manageable absolute sum, and a 20 percent FX move is painful but not insolvency-inducing. The borrower who takes the same trade at JPY 500 million on a luxury Tokyo apartment, financing 70 percent of the property value, is in 2007 Hungarian territory.
The 2026 yen carry mortgage is not, in itself, mispriced. It is correctly priced for a sophisticated buyer who understands they are taking FX risk. It is mispriced for a buyer who believes they are taking interest-rate risk.
The math of when it works and when it kills you
The honest framework for foreign-currency mortgages reduces to a single question: do you have natural income, or expected income, in the currency of the loan?
If yes, the foreign-currency mortgage is a hedge. If no, it is a leveraged short position.
Case A: Natural hedge (works)
A Swiss-based executive earning in CHF buys a Swiss property and finances it with a CHF mortgage. The interest rate is low, the currency exposure is zero (income and debt in the same currency), and the borrower is fundamentally taking only interest-rate and property-value risk. This is a normal domestic mortgage and is uncomplicated.
The same Swiss-based executive who buys a Portuguese property and finances it in EUR against CHF income is taking a small FX position, typically considered acceptable because the property itself is denominated in the loan currency, producing a partial hedge.
Case B: Rental-income hedge (works conditionally)
A US-based investor buys a Dubai property generating AED rental income and finances 50 percent of the purchase with an AED mortgage. The rental income services the AED debt directly. The FX risk is contained to the equity portion. This is a legitimate hedge structure used widely in cross-border investment property.
The condition: the rental income must be reliable enough, in the loan currency, to service the debt through the cycle. If the property is empty for six months, or if rental yields compress, the natural hedge breaks down and the investor must service AED debt from USD income, reintroducing FX risk.
Case C: Pure leverage on a foreign currency view (works only with a thesis)
A buyer who is deliberately taking a directional view that the funding currency will weaken against their earning currency over the loan term may take a foreign-currency mortgage as an expression of that view. This is the only scenario in which a foreign-currency mortgage without a natural hedge is a defensible trade.
The caveat: in this case, the borrower is operating as an FX trader with a 25-year horizon and high leverage. The position size relative to total household balance sheet should be calibrated accordingly, which in practice means small.
Case D: The trap (kills you)
A buyer who earns in their local currency, has no income in the loan currency, has no rental flow in the loan currency, and takes a foreign-currency mortgage because the headline rate is lower than the local-currency alternative. This is the Hungarian household in 2007. This is the Polish frankowicze. This is the unwary 2026 yen mortgage buyer.
The framework: if you cannot articulate the natural hedge, you are in case D.
Country-by-country: 2026 foreign-currency mortgage availability
Eurozone (EUR mortgages for non-eurozone residents)
EUR mortgages are widely available to non-resident property buyers across Portugal, Spain, Italy, France, Germany, the Netherlands, Greece, and Cyprus, generally at LTVs of 60–70 percent and rates that track the ECB main refinancing rate. For a US-, UK-, or Swiss-based buyer purchasing eurozone property, the natural hedge is whether the rental income is in euros. If yes, this is conventional. If no, the borrower is taking FX risk on the EUR/home-currency pair.
Japan (JPY mortgages for non-residents)
JPY mortgages for non-residents are available through a narrow set of banks, Tokyo Star Bank, SMBC Trust Prestia, and certain private-banking arms, generally requiring substantial deposit relationships, LTVs of 50–60 percent, and rates of 1.0–2.0 percent in 2026. The natural hedge is JPY rental income from the property. The trap is a borrower who has no JPY income and is treating the loan as a low-coupon mortgage rather than a leveraged short yen position.
UAE (AED and USD mortgages)
UAE mortgages are available in AED at rates broadly tracking EIBOR and in USD for certain offshore-domiciled buyers at rates tracking SOFR. AED is pegged to the USD, which means the FX risk is, in practice, the dollar's risk against the buyer's home currency. For US, GCC, and dollar-denominated buyers, this is a near-natural hedge. For euro-, sterling-, or yen-earning buyers, it is dollar exposure.
United Kingdom (GBP and USD mortgages for international buyers)
The UK offers GBP mortgages widely to international buyers at LTVs of 60–75 percent through Lloyds International, HSBC International, and a handful of specialist lenders. USD-denominated mortgages on UK property are also available for high-net-worth international buyers through private banks. For a buyer with sterling rental income, GBP is the natural hedge. For a buyer who plans to live in the property eventually and lacks sterling income, USD financing is sometimes used to align with home-currency exposure, at the cost of a more complex two-currency position.
United States (USD mortgages for non-residents)
USD mortgages on US property are available to non-residents through HSBC US, Citibank International, and a number of foreign-national mortgage specialists, typically at LTVs of 60–75 percent and rates 50–150 bps above the equivalent resident product. For dollar-earning international buyers (which is a large segment of the global wealthy), this is straightforward. For non-dollar earners, the USD mortgage is a directional position on the dollar.
Switzerland (CHF mortgages for non-residents)
Despite the historical disaster, Swiss CHF mortgages remain available to non-residents through the cantonal banks and the major Swiss private banks, with strict Lex Koller property-eligibility rules and LTVs typically capped at 50–60 percent. The natural hedge is Swiss rental income (rare for non-resident-permissible properties) or a deliberate franc thesis. Anything else is a textbook FX position.
Turkey (TRY and USD mortgages)
Turkish lira mortgages at headline rates of 30–40 percent are not in practice competitive with cash purchase or with USD-denominated mortgages from Turkish banks' international units. The lira's structural depreciation against hard currencies makes the lira mortgage, for a hard-currency-earning buyer, a poor product even at face. The USD mortgage on Turkish property, where available, is the more common structure, but introduces its own FX risk for non-USD buyers.
Latin America (USD mortgages on Mexican, Panamanian, Costa Rican property)
Mexican property is overwhelmingly financed by foreign buyers in USD through international banks or, for cash buyers, not financed at all. The peso mortgage is structurally similar to the lira mortgage: high rate, depreciating currency, poor product for hard-currency-earning buyers. Panama operates in USD natively (no currency mismatch for dollar earners). Costa Rican property is increasingly financed in USD by international buyers, with the colon mortgage limited to local-income borrowers.
The 2026 hard-currency debt thesis
A quieter trend through 2024–2026: cross-border property buyers in soft-currency jurisdictions, Turkey, Argentina, parts of Africa, parts of Southeast Asia, are increasingly buying property in hard-currency-denominated markets (USD, EUR, CHF, GBP) and financing those purchases with hard-currency mortgages, with rental income in hard currency.
The thesis: rather than try to time the soft-currency cycle from home, lift the entire balance sheet to a hard-currency base. The mortgage is denominated in the same currency as the property and the rental income; the soft-currency earning is converted up-front at the equity stage and never re-enters the equation. The FX risk on the soft-currency side is collapsed to a single conversion event.
This is the inverse of the 2007 CHF mortgage. The 2007 borrower converted hard-currency debt into soft-currency liability. The 2026 hard-currency buyer converts soft-currency equity into hard-currency assets, debt, and income, and the FX exposure of the future household is denominated in the hard currency rather than the soft one.
For Turkish, Argentine, Egyptian, Nigerian, and similar buyers, this is the durable response to the currency-cycle problem that the 2007 CHF trade pretended to solve.
The JanusHermes framework
Before signing any foreign-currency mortgage in 2026, the cross-border buyer should be able to answer four questions in writing.
1. What currency do I earn in, today and prospectively over the loan term?
If the answer changes, for example, you currently earn in USD but plan to retire in five years to a country where you will live on a euro pension, the analysis must reflect the future currency exposure, not the current one.
2. What currency does the property generate income in?
Rental income in the loan currency is the natural hedge. Rental income in your earning currency is not. Vacant primary-residence property generates no income and provides no natural hedge.
3. What is the loan principal in my earning currency after a 20 percent adverse FX move?
Run the math. If the loan principal in your earning currency after a plausible 20 percent funding-currency strengthening exceeds 80 percent of the original property value, you are in 2007 Hungarian territory.
4. What is the loan principal after a 40 percent adverse FX move?
The CHF/HUF move from 2008 to 2015 was approximately 100 percent. The yen has moved 30–40 percent in past cycles. The pound moved 25 percent in 2016. Forty percent is not a tail; it is well within the historical experience of every major currency pair in the last twenty-five years.
If the answer to question 4 is "I lose the property," the foreign-currency mortgage is not for you. If the answer is "I refinance into local currency or sell the property at a manageable loss," the position is sized correctly.
The 2026 cross-border buyer who internalizes these four questions before signing is the buyer who keeps the property through the cycle. The buyer who is sold on the headline coupon is the buyer who, twenty years from now, will be cited in a future version of this article.
Frequently asked questions
What is a foreign-currency mortgage and why is it risky?
It is a loan denominated in a currency you do not earn in, effectively a low-rate loan bundled with a short position on that currency. If the loan currency strengthens against your earning currency, the balance you owe rises in your own currency even though the property's local price has not changed.
What happened with the Swiss franc mortgages in Central Europe?
Roughly 1.5 to 2 million households across Hungary, Poland, Croatia, and Romania took CHF mortgages at 2–3 percent before 2008. After the Swiss National Bank removed its EUR/CHF floor in January 2015 and the franc revalued, many owed far more in local-currency terms on properties that had not appreciated, leading to defaults, government conversion programs, and years of litigation.
Is the 2026 yen mortgage the same trade?
Mechanically, yes. A buyer financing Japanese property with a yen mortgage at around 1.5 percent while earning in another currency is implicitly short the yen. The yen has moved 30–40 percent in past cycles, so a borrower with no yen income is taking FX risk, not just interest-rate risk.
When is a foreign-currency mortgage actually safe?
When you have natural income in the loan currency, either you earn in it, or the property generates reliable rental income in it that services the debt. If you cannot articulate that natural hedge, you are taking leverage on a currency view, which should be sized small.
The FX-stress framework in this guide, rate shock, currency shock, and the natural-hedge test, is the underwriting discipline to apply before signing any foreign-currency mortgage.
A note on the numbers: where no source is named, the market figures in this article (prices, yields, costs) are indicative estimates compiled from publicly available market data and industry reporting at the time of writing. Markets move and rules change, so treat them as a starting point and verify current figures with official sources before acting on them.
