[Yonhap]
[Yonhap]

A man in his 70s, identified only as A, traded the ETF product "T○○○○ Semiconductor TOP 10" a total of 13 times between Jan. 6 and June 4 this year. His average holding period was 10 days, with a cumulative return of 78.8%.

He invested 100 million won ($69,100) in principal, and the evaluated value of his holdings grew to 179 million won. He paid 17.04 million won in upfront fees.

Had A chosen a back-end fee structure under the same conditions, his back-end fees would have been 560,000 won and his portfolio value would have grown to an estimated 203 million won. Choosing back-end fees would have generated a compounding benefit of 7.99 million won on the savings from avoided upfront fees — exceeding the 16.22 million won in fee savings alone — pushing the estimated cumulative return to 103.1%.

ETF trust sales at six domestic banks jumped 8.8 times in five months, the Financial Supervisory Service said. The fee burden on consumers is growing as short-term trading and upfront fee arrangements increase, even as trading by vulnerable investors — including the elderly — rises.

In response, the Financial Supervisory Service (FSS) issued a Caution-level consumer alert Thursday and published guidance on ETF trading precautions to help retail investors choose fee structures suited to their investment horizon and risk profile.

Short-term trading habits clash with fee structures

According to the FSS, the total value of ETF transactions across all financial sectors from January last year through May this year reached 3,290 trillion won, of which bank ETF trust sales accounted for 2.0%, or 66 trillion won.

Although banks hold a small share of overall ETF trust sales, that share has been growing rapidly this year amid a domestic stock market boom. In won terms, sales expanded from 6.5 trillion won in March to 9.2 trillion won in April and 14.4 trillion won in May — a 1.0 percentage point increase in market share over three months.

ETF trading was concentrated at KB Kookmin, Shinhan, Hana, Woori, SC First and NH NongHyup. The six banks recorded ETF sales of 64 trillion won from January 2025 through May 2026, spanning 1.03 million transactions. Sales in May alone reached 10.8 trillion won, jumping from 1.2 trillion won in December last year.

The FSS said bank ETF trust trading shows a strong tendency toward short-term dealing, and identified a serious mismatch between that trading behavior and the trust fee structures customers are being placed in.

An FSS analysis of ETF trust transactions found that the average holding period was just 42 days and the same customer entered into contracts an average of 5.5 times — a pattern that became more pronounced this year.

Yet despite 94.6% of ETF trades being short-term deals sold within six months, upfront fees — which are disadvantageous for short-term trading — accounted for 91.7% of fee arrangements.

More than half of accounts — 58.1% — had target returns set at 5% or below, and those set at excessively low thresholds of 3% or 1% or below accounted for 20.8% and 1.1%, respectively. The FSS said the lower a consumer sets their target return, the more frequently trades occur; when combined with an upfront fee structure, the fee burden rises while lost investment opportunities from fee payments compound the damage.

In practice, the combination of upfront fees and low target return settings allowed banks to collect fees more than seven times the optimal trust fee rate for their customers, the FSS said.

Another notable trend since January last year has been a rise in ETF trust trading by vulnerable investors, including the elderly and those with no prior equity investment experience.

During the period, investors aged 65 and older accounted for 29.9% of traders, and those 80 and older for 1.1%; the monthly share of ETF buyers aged 65 or older showed a clear upward trend. Investors with no prior equity investment experience made up 1.5% of the total, and new contracts from that group reached 1,426 in May this year — 2.2 times the 657 recorded in December last year.

Excessively low target returns drive up fee burden

The FSS said that as a general rule, upfront fees are more advantageous when the investment period is one year or longer, while back-end fees are better for periods of one year or less. It particularly advised investors expecting short-term trading to choose back-end fees, noting that the shorter the investment period, the more favorable back-end fees become.

The FSS also warned that setting target returns at excessively low levels — such as 1 to 3 percent — can trigger frequent trading and cause the fee burden to outweigh the returns earned.

The regulator also cautioned that bank ETF trusts may not be suitable for repeated short-term trading, as they carry higher transaction costs — including trust fees — compared with direct trading through a securities firm.

The FSS added that even when purchased through a bank, ETF trusts do not guarantee the principal, and investors should note that ETF trusts through banks cannot be traded in real time. Banks execute ETF orders in split or delayed transactions during trading hours after receiving a customer's instruction, meaning the actual execution price can differ significantly from the price at the time the instruction was given when share prices move sharply.

The FSS plans to form a task force with industry groups and associations to discuss the trust fee structure covering ETFs, bank internal performance evaluation metrics and improvements to the sales process.

On the fee structure in particular, the regulator intends to review everything from scratch — including upfront and back-end fees, early redemption fees and trading commissions.


ehkim@heraldcorp.com