South Korea's Jeonse System: When the Tenant Finances the Landlord

Jeonse looks like a rental contract without monthly rent. Economically, it is also credit: the tenant transfers a large deposit to the landlord, who must return it when the lease ends. If the tenant borrows to assemble the money, the same transaction creates a bank debt for the tenant and a repayment liability for the landlord. If the landlord fails to pay, a public guarantee can move the risk to HUG, which compensates the tenant and then tries to recover its claim. At the end of 2025, HUG guaranteed KRW126.7887 trillion of jeonse-deposit refunds. That stock is not a forecast loss. It reveals the scale of a system in which a rental agreement also operates as a housing-finance chain.

Jeonse is both a lease and a loan to the landlord

Under a pure jeonse contract, the tenant pays a large lump sum at the beginning of the lease and makes no monthly rental payment. The Korea Development Institute defines jeonse as “lump-sum deposits with no monthly payments.” The landlord has use of the funds during the contract and must return the deposit.

The arrangement does not make the tenant an owner of the home. It creates a refund claim against the landlord. For the tenant, the deposit is an immobilized asset. For the landlord, it is a liability falling due at the end of the lease. The absence of a visible monthly rent does not make housing free: the cost may take the form of foregone income on savings, interest on a jeonse loan, or a combination of the two.

The IMF’s 2023 Korea report described the deposit as a large interest-free transfer, often around 50% to 70% of the home’s value, made in lieu of monthly rent. That range is a broad historical description of the market, not a legal rule or the ratio for any particular contract.

Korean law supplies a reference horizon. Article 4 of the Housing Lease Protection Act, in force from 2 January 2026, generally deems a lease with no term or a term shorter than two years to run for two years, while allowing the tenant to assert the shorter agreed term. More importantly, it treats the lease relationship as continuing until the deposit is returned.

A lease without monthly payments does not eliminate housing cost

The economic cost of jeonse depends on where the funds come from. A household using savings gives up the return those funds could have earned. A household borrowing the money pays interest to a bank. In either case, the landlord receives capital that can be used subject to applicable law, without paying the tenant a bond-like coupon.

Not every jeonse tenant borrows. The loan market is nevertheless official and important enough for HUG to offer a combined product. HUG’s Jeonse Safe Loan Guarantee joins two different protections: a deposit-refund guarantee for the tenant and a loan-repayment guarantee for the financial institution. The loan guarantee generally covers 80% in the capital region and regulated areas and 90% elsewhere, subject to product ceilings and eligibility rules.

One lease can therefore connect four positions:

• the tenant’s claim on the landlord;
• the tenant’s possible debt to a bank;
• the landlord’s liability to the tenant;
• HUG’s contingent guarantee.

This layering explains why jeonse can increase housing leverage without looking like a conventional mortgage. Part of the home’s financing is embedded in the lease itself.

Refinancing can arrive in the form of the next tenant

When the lease ends, the landlord can repay from cash, sell or refinance the home, borrow elsewhere, or use the next occupant’s deposit. There is no basis for assuming that every landlord follows the last method. It does create a straightforward rollover risk: if the new deposit is smaller than the old one, the difference must be funded.

This is known as reverse jeonse. Contracts signed when deposits were high can mature after market deposits have fallen. The landlord then owes the departing tenant more than can be obtained from the incoming tenant.

The IMF’s 2023 Article IV report on Korea cited a June 2023 Bank of Korea estimate: the potential jeonse-deposit gap on contracts maturing in 2023 and 2024 was about KRW44 trillion, or 2% of GDP at the time. It was an estimate of a funding need, not a realized loss.

The direction of risk changes with the market. Falling deposits reduce the entry cost for a new tenant but can open a liquidity hole for the landlord. Rising deposits facilitate the landlord’s rollover while requiring more savings or debt from the next household.

In June 2026, the KDI reported that jeonse prices rose 0.38% month on month, while housing prices increased 0.33%. This is a monthly national index, not a measure of risk in each contract. It does show how the system can rebuild the incoming tenant’s upfront burden after the phase of falling deposits that weakened landlord repayments.

The home’s value is not enough: claim ranking matters

The deposit is protected by formalities and statutory rights, but it does not automatically rank ahead of every other claim.

Article 3 of the Housing Lease Protection Act provides that delivery of the home and resident registration make the lease effective against third parties from the following day. Article 3-2 adds that a tenant who satisfies those requirements and obtains a fixed date can receive payment from auction or public-sale proceeds ahead of junior right-holders and other creditors.

“Junior” is the key word. The date of a mortgage, the fixed date on the lease, certain tax claims, other tenants and special rules for small deposits can change the actual ranking. A sale price that looks sufficient may still leave the tenant short if higher-ranking rights absorb the proceeds first.

HUG applies that logic in its underwriting. Its deposit-refund guarantee generally caps coverage at 90% of the recognized home value minus senior claims. For detached or multi-household homes, senior claims can include deposits held by other tenants who rank ahead of the applicant.

That calculation does not ensure the recognized value will be obtained in an auction. It is an admission barrier for the guarantee. Forced-sale discounts, costs and newly identified priority claims remain legal and economic variables.

HUG turns a landlord default into a public claim to be recovered

The Korea Housing & Urban Guarantee Corporation, or HUG, is a public corporation established under Article 16 of the National Housing and Urban Fund Act. Its corporate overview lists jeonse-deposit refund guarantees among its public-policy guarantee activities.

The product does not automatically cover every lease. Under the terms published by HUG, the eligible deposit cannot exceed KRW700 million in the capital region and KRW500 million elsewhere. An application generally has to be submitted before half of the contract term has elapsed. Coverage remains subject to the recognized home value, senior claims and the other product conditions.

Two main events can support a claim: the deposit remains unpaid one month after termination or expiry without a valid reason, or an auction leaves the tenant unable to recover the full amount. HUG’s claims process sets out notification, claim submission, review and subrogation. HUG says payment is made within one month of an eligible claim and generally requires the tenant to vacate the property.

Subrogation changes the creditor, not the economic debtor. HUG pays the tenant and then holds a claim against the landlord and other responsible parties under the case. It can investigate assets, preserve claims, pursue legal action, sell a claim or ultimately recognize an unrecoverable loss.

The protected tenant regains liquidity more quickly. The risk has not disappeared: it has moved to the guarantor’s balance sheet and is then reduced, or not, by recoveries.

Three HUG figures that must never be added together

HUG’s disclosure pages make it possible to separate three measurements.

At the end of December 2025, the outstanding balance of jeonse-deposit refund guarantees was 1,267,887 units of KRW100 million, or KRW126.7887 trillion. HUG’s total guarantee balance was KRW666.5366 trillion. The first number is a stock of contingent commitments still outstanding, not expected payments.

During 2025, HUG issued 649,784 units of KRW100 million in jeonse-deposit refund guarantees, or KRW64.9784 trillion. This is an annual flow of guarantees issued, not the year-end stock.

Finally, HUG paid 17,935 units of KRW100 million under this guarantee in 2025, or KRW1.7935 trillion. The disclosure page says execution amounts are measured as subrogation payments.

The KRW126.7887 trillion balance represented about 19% of all HUG guarantees at end-2025, an l0g calculation. That share measures the product’s weight in the guarantor’s portfolio. It measures neither default probability nor loss severity.

The shock peaked in 2024 and fell sharply in 2025

The official payment series changes scale after 2022. Jeonse payments increased from KRW0.9241 trillion in 2022 to KRW3.5544 trillion in 2023 and KRW3.9948 trillion in 2024. They then fell to KRW1.7935 trillion in 2025.

The 2025 decline was 55.1% year on year, calculated from HUG data. That is a clear improvement. Payments nevertheless remained 1.94 times their 2022 level. Across 2023 to 2025, HUG paid a cumulative KRW9.3427 trillion under this guarantee alone.

These payments include covered non-refunds and auction shortfalls. They should not be treated as a count of jeonse fraud alone: a guarantee event can arise from inability to pay without fraudulent intent.

A HUG payment becomes a claim, not necessarily a loss

After paying, HUG tries to recover the money. Its debt-management disclosure defines subrogation as performance of the guaranteed obligation in place of the debtor after a guarantee event, and recovery as collection of the paid principal and delay compensation through asset investigation, preservation measures and legal proceedings.

Across all products, the page reports KRW3.4840 trillion of subrogation payments and KRW2.1766 trillion of recoveries in 2025. Those figures must not be subtracted to calculate a “2025 loss.” Recoveries during the year may relate to older payments, and the series covers every HUG product rather than jeonse alone.

A second precaution is necessary. HUG’s guarantee-execution page reports a 2025 all-product total of KRW3.5341 trillion, compared with KRW3.4840 trillion on the debt-management page. The visible notes on the two official pages do not reconcile the KRW50.1 billion difference. This analysis therefore uses:

• the jeonse-specific row on the execution page for jeonse payments;
• the debt-management page only to explain aggregate subrogation and recovery;
• no “net” constructed by mixing the two tables.

HUG’s balance sheet recovered after two years of heavy losses

HUG’s summary accounts cover the whole company. They show a net loss of KRW3.8598 trillion in 2023 and KRW2.5198 trillion in 2024. In 2025, HUG returned to a KRW1.5749 trillion net profit.

At end-2025, HUG reported KRW9.1 trillion in assets, KRW7.4718 trillion in equity and KRW1.6282 trillion in accounting liabilities. The outstanding jeonse guarantee balance was therefore about 17 times equity, an l0g calculation. This is not a regulatory ratio or a loss forecast: a contingent guarantee is not a 100%-payable liability, and severity depends on defaults, recovery rates and correlation among claims.

Loss-absorption capacity was also strengthened through capital increases. HUG’s official history records paid-in capital of KRW4.4151 trillion in February 2024, KRW8.4151 trillion in March 2024, KRW9.2494 trillion in July 2025 and KRW9.6494 trillion in September 2025.

The pages used here do not break down 2025 net income by product or provide the detailed source of each capital increase. It would therefore be excessive to attribute the entire return to profit to lower jeonse payments or to describe every capital increase as a final budget loss.

The public-risk channel is more precise. HUG is a public corporation issuing guarantees in support of housing policy. When defaults rise together, its liquidity and capital absorb the shock before recovery. In 2023, the IMF estimated that under severe scenarios covering several guarantee companies and end-2022 exposures, implied fiscal contingent liabilities remained limited to about one quarter of one percent of GDP. That historical result is neither a 2026 HUG stress test nor proof that every future scenario would be contained.

Jeonse creates leverage that bank-credit statistics do not fully display

A tenant’s bank debt is visible in the credit system. The landlord’s refund obligation is a private liability embedded in the lease. Both may indirectly finance the same housing asset, but they do not necessarily appear under the same statistical heading.

At end-June 2026, the Bank of Korea reported KRW2,019.8 trillion in household credit, including KRW1,891.3 trillion in household loans. The total covers much more than jeonse: mortgages, other loans and merchandise credit. It provides the broad leverage context, not a measure of jeonse loans.

A Bank of Korea working paper released on 3 August 2026 uses jeonse precisely as a capitalized price of housing services. With monthly district data and a panel covering 710,000 borrowers per quarter, the authors distinguish housing booms dominated by demand for housing services from those dominated by asset valuation.

Their result is macroeconomic: valuation-dominant episodes are associated with more household borrowing and lower subsequent spending, especially among borrowers whose debt had previously grown more quickly. The paper does not establish that jeonse alone causes the consumption decline. It shows why the system is analytically powerful: it helps separate the value of using a home from the asset-valuation component.

The hidden jeonse balance sheet therefore matters beyond the rental market. When the deposit or debt required to finance it increases, more household resources are immobilized or devoted to debt service. When a landlord must fill a refund gap, the landlord’s own capacity to consume, invest or borrow deteriorates.

Four scenarios, without a crisis prophecy

1. Deposits fall again

Incoming tenants need less capital, but landlords who received high deposits must fill the gap at maturity. Stronger landlords use cash or refinance. Weaker ones delay repayment, sell or trigger a guarantee. A fall in the national index does not reveal the distribution of risk: location, property type, senior debt and the tenant’s rank all matter.

2. Deposits rise quickly

Rollover becomes easier for some landlords. Incoming tenants need more savings or debt. Immediate repayment risk declines while the new household’s leverage and the size of the future refund claim increase.

3. The home loses value while senior claims remain high

The safety margin narrows. A forced sale may not cover the deposit after better-ranked claims and costs. An unguaranteed tenant bears the shortfall according to legal rights; HUG pays an eligible guaranteed tenant under the contract and then seeks recovery from the landlord.

4. Claims become correlated

A guarantee works comfortably when defaults remain dispersed. A broad housing decline can affect landlords exposed to the same prices and refinancing conditions at the same time. The guarantor’s cash outflows then accelerate before recovery proceedings finish. That is how a lease risk becomes a public-balance-sheet risk.

The mechanism differs from developer finance, but it complements our analysis of Chinese real estate as a chain of claims and guarantees: in both cases, the decisive question is which balance sheet absorbs the gap when refinancing is no longer enough.

What now deserves monitoring

The first indicator is the gap between deposits owed on maturing contracts and those available on comparable new leases. A national jeonse index gives direction, but risk emerges contract by contract and district by district.

The second is the relationship among recognized home value, deposit and senior claims. HUG’s own underwriting uses this structure. Higher home prices improve the buffer; simultaneous increases in deposits or senior debt can consume it.

The third is the sequence of outstanding guarantees, issuance, payments, recoveries and losses. A stable balance with lower payments tells a different story from a rapidly rising balance and delayed recoveries. The summary HUG pages used here do not publish a complete vintage matrix matching each jeonse payment with its ultimate recovery.

The fourth is guarantor capacity: equity, paid-in capital, earnings, liquidity and underwriting rules. The 2025 return to profit reduces visible pressure, but it does not replace an analysis of the contingent risk embedded in KRW126.7887 trillion of jeonse guarantees.

The fifth is household debt and its effect on spending. Aggregate Bank of Korea data cannot isolate the whole jeonse chain. They do show that it operates inside an already highly leveraged household sector.

What the evidence supports

Jeonse is neither a folkloric anomaly nor simply a generous lease without monthly payments. It is a hybrid contract exchanging a rent stream for repayable lump-sum financing. It distributes risk among the landlord, tenant, possible bank, other creditors and guarantor.

The repayment shock eased sharply in 2025: HUG’s jeonse payments fell 55.1% and the corporation returned to profit. Normalization remains incomplete. Payments were still almost twice their 2022 level, the guarantee balance reached KRW126.7887 trillion, and HUG continues to manage claims originating in prior years.

The most important conclusion is an accounting distinction. HUG does not erase the landlord’s debt. It first converts it into liquidity for the tenant and then into a claim held by a public corporation. System quality therefore depends as much on underwriting and legal ranking as on housing prices.

Jeonse can remain stable when deposits, home values and landlord resources evolve without discontinuity. It becomes fragile when the next tenant’s capital acts as the refinancing variable for an old liability. Behind the key to an apartment lies a classic finance question: who provides the funding, who carries the maturity, and who absorbs the loss when collateral value is no longer enough?

Sources and methodology

This analysis is current to 30 August 2026. It separates published data, l0g calculations and hypothetical illustrations.

• Korea Development Institute, August 2026 Economic Bulletin, for the jeonse definition and June 2026 indices.
• Bank of Korea, Good Housing Booms, Bad Housing Booms, 3 August 2026.
• Bank of Korea, Household Credits in Q2 2026.
• Housing Lease Protection Act, version effective 2 January 2026, particularly Articles 3, 3-2 and 4.
• HUG corporate overview.
• HUG jeonse-deposit refund guarantee and claims process.
• HUG combined jeonse-deposit and loan guarantee.
• HUG disclosures for guarantee issuance, payments, debt recovery, financial results and outstanding balances.
• HUG history of capital increases.
• IMF, Republic of Korea 2023 Article IV Consultation.

HUG tables use 억원, units of KRW100 million. The conversions use no exchange rate: 10,000 units equal KRW1 trillion. The 19%, 55.1%, 1.94 and approximately 17-times ratios are l0g calculations from official data. The 100, 80 and 110 indices in the refinancing diagram are purely pedagogical.