Korea Electric Power Corp. (Kepco) tried to turn future electricity bills into money it could use now. Its proposal called for Samsung Electronics and SK hynix to advance a combined ₩25 trillion to help build power networks for semiconductor clusters in Yongin and Honam. Reports on September 14 said both companies had rejected the plan; a Kepco spokesman subsequently confirmed the rejection.
3
37
Why Kepco wanted five years of bills early
The proposed split was ₩20 trillion from Samsung and ₩5 trillion from SK hynix, based on roughly five years of projected electricity bills. Kepco wanted to finance transmission construction while reducing its reliance on new bonds.
3
4
That need sits alongside a longstanding mismatch between Kepco’s costs and the prices it can charge. International fuel-price increases were not fully reflected in electricity bills, contributing to substantial losses and borrowing. Even after the immediate energy crisis, electricity rates remained too low to fully cover supply costs, according to reporting on Kepco’s finances.
14
25 By the end of June 2026, its total debt stood at a reported ₩210.7 trillion; one report put daily interest expense at about ₩11.5 billion.
11
Tariff decisions also limit how quickly Kepco can rebuild its finances. For the third quarter of 2026, the government and Kepco kept the fuel-cost adjustment at +₩5 per kilowatt-hour, rather than changing that component of the bill.
17 Advancing payments could ease a cash shortage, but it would not make electricity sales more profitable.
How the rejected arrangement was meant to work
The chipmakers would have paid for electricity ahead of use. Kepco would then deduct future charges from their prepaid balances and credit interest, with reporting placing the proposed return above the yield on two-year South Korean government bonds.
13 Another account described a rate between government-bond and Kepco-bond yields, but no final rate was agreed.
35
“Five years upfront” did not necessarily mean a single payment on day one. One report said Kepco had floated installments of ₩2 trillion through the end of 2027.
11 The payment schedule and detailed financial terms never became an executed agreement. The companies reportedly judged that tying up so much cash was difficult to justify amid uncertainty about the semiconductor cycle and their own investment needs.
9
13
Why the bond deadline matters—and what remains
Kepco’s temporary, expanded statutory bond-issuance limit is reported to expire at the end of 2027. That makes alternatives to issuing more bonds especially relevant as grid spending continues.
10
27
The ₩25 trillion should not be described as an automatic ₩25 trillion reduction in outstanding bonds. If Kepco had used every won to repay bonds, the arithmetic reduction could have been that large. If it had spent the money on transmission construction, the benefit would instead have been bonds it might not need to issue. Its reported total debt is not a measure of outstanding bonds.
3
11
Other routes have different trade-offs. Bank loans or commercial paper could supply cash but would still require repayment. Extending the bond ceiling could preserve borrowing capacity without fixing the gap between tariffs and costs. Government equity would add capital: a September report said the government had included ₩500 billion in cash investment for Kepco in its proposed 2027 budget, while further capital-raising and bond issuance were under discussion.
27 Public funding for policy-driven electricity discounts could also shift those costs away from the utility, but the provided reporting does not establish an approved amount or plan.
The rejected prepayment might have helped Kepco start grid work sooner and issue fewer bonds. It would not have guaranteed grid-connection dates or paid for all future construction. Above all, it would have exchanged cash received today for bills collected over the following years, with interest owed along the way. Kepco still needs a way to finance new infrastructure while covering the cost of supplying power.
13
14