- Published 2026.11.10 11:47
Six online lenders’ personal loan delinquency rate at 1.09% as of September
Denominator effect from 36-fold surge in monthly originations in just over a year
Whether delinquency can be contained as new lending slows and maturities arrive
Online investment-linked finance companies (online lenders) have posted a personal unsecured loan delinquency rate in the 1% range on the back of financial institution-linked investments, but some point out that this is a short-term illusion caused by the explosive growth in originations. As loan assets have expanded rapidly in a short period, the credit risk is merely masked by the denominator effect, and the actual asset soundness is expected to be determined when the loans mature.
According to data released by the Online Investment-Linked Finance Association on Nov. 10, the proportion of private mid-rate loans at six online lenders participating in financial institution-linked investments reached 94.8% as of the end of September. The delinquency rate, including loans overdue by 30 days or more and individual rehabilitation claims, stood at 1.09%. Excluding individual rehabilitation claims, the rate is 0.76%. This is ostensibly lower than the delinquency rates of mid- to low-credit household unsecured loans at the three internet-only banks as of the end of May: K bank at 1.79%, KakaoBank at 2.16%, and Toss Bank at 3.07%. The online lenders included in this tally are MoneyMove, Mouda, Earnest AI, 8Percent, PFCT, and Hanpass Finance.
However, from a credit risk analysis perspective, it is difficult to take this figure at face value. The biggest reason is the abnormal rapid growth in asset balances. The six companies’ monthly new personal unsecured loan originations surged about 36-fold from 2.75 billion won (about $2 million) in May 2025, when financial institution-linked investments began, to 99.94 billion won (about $73 million) in September this year. Cumulative originations reached 811.4 billion won (about $596 million). When the denominator—total loan balance—expands rapidly in a short period, the delinquency rate falls sharply due to the time lag before actual defaults emerge in new loans, creating an illusion.
The borrower composition and loan characteristics also carry potential credit risk. The average credit score of borrowers in this tally, based on Korea Credit Bureau (KCB), is 686, and loans to borrowers with credit scores of 675 or below—the bottom 10%—accounted for 45.0% of supply. A significant portion of the loans are refinancing loans for multiple-debt borrowers who consolidate card loans or lending business loans. Immediately after refinancing, the repayment burden is reduced and borrowers continue to make normal payments, but these borrowers are inherently high-risk and vulnerable to economic slowdowns or employment shocks.
The timing of bond maturities is also considered a significant risk factor. Most loans were originated intensively within the past year or so, and the proportion of bonds where borrowers have fully repaid principal is extremely rare. In the early stages of a loan, borrowers pay only interest or a portion of principal and interest, so signs of distress remain below the surface. The actual performance of the artificial intelligence (AI)-based alternative credit scoring models touted by the online lending industry can only be proven 18 months to two years after loan origination, when the repayment burden peaks.
Behind the current 1% delinquency rate lies institutional limitations. If a macroeconomic shock materializes into defaults, institutional investors’ funding supply could slow and the expansion of new originations could halt. At that moment, the latent delinquency amounts would fully surface in the numerator, and there is a risk that the delinquency rate would rise sharply. Whether this soundness can be maintained when new origination volume enters a stagnation phase and the refinancing loans’ full repayment maturities arrive will be the true test of the online lending sector’s credit assessment capabilities.
(Dollar figures are approximate, converted automatically at 1,000 won = $0.735, the rate on Nov. 10, 2026.)
※ This article is an AI translation of the Korean original published by Korea Credit News; it was checked by automated quality guards and published by the English edition desk.
