P2P Lending in Korea: How It Works for Firms & Investors

Korean P2P lending — legally, online investment-linked finance (온라인투자연계금융, “온투업”) — is routinely misunderstood on both sides of the transaction. It is not a high-interest savings substitute for investors, and it is not just “faster bank credit” for borrowers. The governing statute defines the structure precisely: an investor’s money is linked to a loan made to a specific borrower, and the right to receive that loan’s principal and interest belongs to the investor.

Three caps define retail P2P investing

Statutory limits under the Online Investment-Linked Finance Act

Per borrower₩5M

Per real-estate loan₩10M

Total invested₩30M

That one sentence explains most of what follows. A bank takes deposits and lends off its own balance sheet — depositors are not exposed to any single borrower. A P2P platform, by contrast, is fundamentally a matchmaker that screens, services, and collects. The investor is not depositing money with the platform; the investor is taking the credit risk of an individual borrower. If that borrower fails to repay, the loss lands on the investor — and the Financial Services Commission explicitly labels these products high-risk with no principal protection.

This guide looks at the structure from both seats: a company (including a foreign-owned Korean company) considering P2P as a borrower, and an individual or corporate investor considering it as an asset. If you are new to Korean finance generally, start with opening a Korean bank account and how loans work for foreigners — this article assumes that groundwork.

A note for foreign readers: the law does not exclude foreigners, but platform onboarding runs on Korean identity verification — in practice an ARC, a Korean bank account, and a Korean phone number. Whether a given platform accepts foreign individual investors, or a foreign-owned 법인 as a borrower, varies; confirm with the platform before you plan around it.

For borrowers: not a bank substitute — capital on different terms

If you run a company in Korea, the right lens for P2P borrowing is not the interest rate. It is three other questions: how fast can the money arrive, what collateral structure does it require, and how does it interact with your existing bank covenants?

Banks underwrite conservatively and comprehensively — credit rating, financial statements, debt-service coverage, existing borrowings, collateral capacity, the CEO’s personal credit — and the process can take weeks. P2P platforms each run their own models, and many weight the recovery value of a specific asset (real estate, receivables, inventory, contracted revenue) more heavily than the overall balance sheet. That makes P2P plausible for a company whose bank credit lines are fully drawn, or which needs bridge money on a timeline banks cannot meet.

A concrete shape: a company has a confirmed ₩1B asset sale closing in three months but needs ₩200M of raw-material purchasing today. A new bank facility would take a month-plus to underwrite and the working-capital line is already in use. A six-month collateralized P2P loan can bridge that gap. The decision variable is not the quoted 10% annual rate — it is the all-in annualized cost after platform fees, collateral registration costs, and any early-repayment charges, weighed against the actual usage period and the risk that the sale slips.

Two hard warnings:

• P2P is not a way around your bank. If your existing loan agreements contain negative covenants — restrictions on additional borrowing, prohibitions on further security interests, financial-ratio maintenance — a P2P loan can breach them just like any other debt. Registering a junior mortgage without re-reading your bank documents can trigger an event of default (기한이익 상실) on the debt you already have.
• P2P works as a bridge, not as a lifeline. It is most efficient when the term is short and the repayment source is identified in advance. The moment a company starts rolling high-rate P2P money to repay existing debt with no clear repayment source, it is no longer financing strategy — it is extending a liquidity problem at premium prices.

For investors: credit exposure without bond liquidity

For individuals with spare KRW — including expats settled in Korea — and for corporate treasuries, the honest classification of P2P investment is illiquid single-name credit, not a deposit alternative.

A quoted 8–12% yield is not an expected return. Subtract expected losses, delinquency drag, recovery costs, platform fees, and tax. On a 10% product, a permanent loss of 3% of principal plus a six-month recovery delay produces a realized return that looks nothing like the marketing page.

Investor limits are set by law

The enforcement decree caps what a general individual investor can deploy:

Cap	Amount
Per single borrower	₩5,000,000
Across real-estate-collateral loans	₩10,000,000
Total connected investments	₩30,000,000

Corporations and professional investors are regulated separately — the decree carves out corporate and institutional connected investment rather than applying retail caps. But a corporate badge does not shrink the risk; it usually enlarges the ticket. A corporate treasury investing in P2P needs its own internal per-project limits precisely because the legal ones do not apply.

Tax treatment differs between individuals and corporations

Do not compare returns using one withholding rate. Interest an individual earns through an FSC-registered platform can qualify for the 14% withholding rate applied to standard interest income, while interest paid to a domestic corporation on non-business loans (비영업대금의 이익) is subject to a 25% withholding interpretation under the Corporate Tax Act — and a corporation’s real tax burden is settled in its annual corporate tax return alongside everything else. Confirm the treatment with your accountant before investing, not after.

The most dangerous number is not the delinquency rate — it is how it is calculated

The most common evaluation mistake is comparing platforms on their published delinquency ratios.

A platform showing 2% is not automatically safer than one showing 5%. Three mechanical distortions make the headline number almost meaningless on its own:

• Growth dilution. When new originations grow fast, the denominator inflates and the ratio falls, regardless of underwriting quality.
• Extension accounting. Loans whose maturities were extended may be reported as current.
• Sale and removal. Delinquent loans sold off to collection firms can exit the statistics entirely.

What to examine instead: cumulative defaults against cumulative originations, the current outstanding balance of delinquent loans, the share of 30-day and 90-day-plus delinquencies, the average time from default to completed collateral disposal, and the final loss rate after recovery. For real-estate project finance and junior-lien products, expect a year or more between a default and actual cash recovery.

Also separate two facts that get conflated: “the loan is delinquent” and “the principal is lost.” A well-collateralized loan with little senior debt can recover most of its value — slowly. Conversely, a platform with a pristine delinquency ratio can hand investors a large loss from a single project whose collateral was overvalued.

“Secured” means nothing — seniority means everything

Never assess a product from the phrase “real-estate collateral” alone.

What matters is the realistic collateral value and your position in the queue. Take a property appraised at ₩2.0B carrying a ₩1.4B senior bank loan, with the P2P platform registering a ₩400M junior mortgage. Nominal LTV: 90%. If the property sells at auction for 75% of appraisal — ₩1.5B — the junior tranche’s recovery is severely impaired before costs are even deducted.

For mortgages, the registry (등기부) priority is the core fact, and auction-distribution claimants that outrank you — certain taxes, wage claims, tenant deposits — need checking too. For receivables or deposit pledges, the words “pledge registered” matter less than whether the underlying claim actually exists, whether the third-party debtor was notified or consented, and whether set-off could erase it.

The law is on your side here, if you use it: platforms are required to disclose collateral value, valuation method, and how the security is registered, plus the collection and debt-sale procedures and costs that apply on default. That disclosure section — not the banner yield — is the part of the listing worth reading.

Choose platforms on underwriting and recovery, not yield

Operating a P2P platform requires FSC registration, with minimum capital tiered at ₩500M, ₩1B, or ₩3B depending on the linked-loan balance, plus staffing, systems, and internal-control requirements. Registration is the floor, not a safety endorsement — it tells you the platform is legal, not that its products are sound.

Four practical screens:

1. The platform’s own financials. Repeated operating losses and fast-eroding equity undermine the platform’s ability to manage a loan book for years — which is what servicing and collection require.
2. Underwriting staff. A platform selling real-estate products with no real-estate-finance or appraisal expertise on the team is a pass at any yield.
3. Growth rate of the loan book. Unusually fast origination growth and loosening credit standards tend to travel together.
4. Behavior after default. How transparently does the platform disclose auction filings, security enforcement, debt sales, and workout negotiations? A platform that leads with yields and rewards while staying vague about recovery procedures has its priorities inverted — and the FSC has repeatedly warned about operators dangling excessive rewards or implied loss protection.

Diversify risk factors, not product counts

Twenty investments are not twenty diversifications. If all twenty are junior liens on Seoul officetels, you effectively hold one bet on one property market — even across different platforms, they share the same appraisal conventions and the same real-estate cycle.

Real diversification cuts across borrower, collateral type, region, maturity, platform, and repayment source. A workable internal rulebook for, say, ₩300M of investable cash: no more than ₩30M to any single borrower, no more than 20–25% of the total on any single platform, and a deliberate mix of recovery structures (real-estate-secured, personal credit, business receivables) rather than a big count of similar products.

For corporate treasury money the bar is higher still. Funds earmarked for payroll, VAT, corporate tax, or supplier payments do not belong in P2P at all — a contractual maturity date is not a guarantee the borrower performs on it. Only surplus cash with no planned use for at least six to twelve months should even be considered. (If you are structuring a Korean company in the first place, see registering a business in Korea.)

The working checklist

• Borrowers: check your existing lenders’ additional-borrowing and negative-pledge covenants first.
• Investors: compute net recovery value — appraised value minus senior claims — not the headline collateral figure.
• Judge net yield after tax, fees, and expected loss, not the quoted rate.
• Weigh 90-day-plus delinquencies and final recovery rates over the published delinquency ratio.
• Set internal caps per borrower and per platform before the first won moves.

P2P finance is not good money because it is faster than a bank, and not a good investment because it out-yields a deposit. It is a legitimate tool for borrowers with a short, defined funding gap and an identified repayment source, and for investors willing to read collateral structures and recovery procedures themselves. Used on headline numbers alone, it becomes the most expensive money either side ever touches.