Why Korean banks keep earning easy money on spreads

Korean banking has an unusually simple income statement. Financial Supervisory Service data show that domestic banks earned interest income of roughly 59 trillion won in 2023, a record, and that net interest income accounts for the large majority of operating revenue at the four financial groups that dominate the sector. Fee businesses — wealth management, investment banking, payments — exist but remain secondary. The core activity is taking deposits and making loans, and the profit is the difference.

That difference widens predictably when policy rates rise. Korean deposit accounts are heavily weighted toward low-yielding demand balances, while a substantial share of household lending is priced off short-term benchmarks that reprice quickly. When the Bank of Korea raised its policy rate from 0.5 percent in 2021 to 3.5 percent by early 2023, loan rates followed within months while deposit rates lagged, and net interest margins expanded across the sector. The banks did not become better at anything. The rate cycle simply moved in their favour.

The public response was sharp enough to become a recurring feature of Korean economic debate: the accusation of “interest business”, the charge that banks profit from a regulated, protected position rather than from competence or risk-taking. Proposals for a windfall levy on excess interest income have been raised in the National Assembly more than once. What emerged in practice, in late 2023, was a negotiated arrangement in which the banking sector committed a package worth around 2 trillion won to interest rebates and support for small business borrowers — a settlement that acknowledged the political problem without changing the structure that produced it.

That structure has three components. The first is concentration: four large financial holding groups, plus a handful of specialised state-linked banks, account for most of the market, and licensing makes new entry rare. The second is the loan book’s composition. Korean bank lending is dominated by mortgages and jeonse deposit loans, collateralised against residential property, with credit losses historically low and capital charges modest — a business that yields steady returns without requiring the credit judgment that lending to small firms or early-stage companies demands. The third is deposit stickiness: Korean households move current accounts rarely, so competition for the cheapest funding is weaker than headline rate comparisons suggest.

Internet-only banks were the deliberate policy answer. K Bank and Kakao Bank were licensed in 2017 and Toss Bank in 2021, with mandates to compete on price and to serve borrowers with thinner credit files. They have delivered on some of this: deposit and loan rate comparison became a consumer expectation, app-based account opening compressed the incumbents’ branch advantage, and Kakao Bank alone reported more than 20 million customers by the mid-2020s. But their combined share of total banking assets remains small, and their own lending has gravitated toward the same low-risk collateralised products, because the economics that favour mortgage lending apply to newcomers too.

Regulators have added other pressure points — open banking infrastructure allowing account aggregation across institutions, mandatory disclosure of deposit and lending rates for comparison, and platform-based loan shopping. These lower search costs at the margin without altering the fundamental balance between four large incumbents and a household sector whose main financial decision, the mortgage, is made once a decade.

There is also a risk dimension that the profitability argument tends to obscure. A banking system whose assets are overwhelmingly claims on residential property is highly correlated with a single asset price. Bank of Korea financial stability reporting has flagged household debt above 90 percent of GDP for years, among the highest ratios in the OECD. The spread income that draws criticism in good years is, in that sense, compensation for a concentration that has not yet been tested by a serious housing downturn.