▲ Stock Trading
Seohak ants, a term referring to retail investors in South Korea investing in the U.S. stock market, have shifted from selling off U.S. stocks earlier this month to a net-buying stance.
They have purchased nearly 2 trillion won worth of U.S. stocks over the past four days, turning this month's net sales into net purchases.
According to SEIBRO, the securities information portal of the Korea Securities Depository, on September 19, local retail investors net-purchased $1.314 billion ($1.8199 trillion) worth of U.S. stocks over a four-day period from September 14 to 17 (based on inquiry dates).
Total purchases were tallied at $3.53943 billion, while total sales stood at $2.22543 billion.
As a result, Seohak ants recorded $1.18299 billion in net purchases from September 1 to 17.
They had posted $131.01 million in net sales through September 11, but switched to net buying following heavy purchases over the four days this week.
From September 14 to 17, Seohak ants heavily bought the Direxion Daily Semiconductors Bull 3X Shs ETF (SOXL), which tracks three times the daily return of the Philadelphia Semiconductor Index.
During this period, net purchases of SOXL reached $823.51 million (1.1411 trillion won), accounting for 62.6% of total net purchases over the same period.
This amount exceeded the net sales of SOXL ($677.44 million) recorded from September 1 to 11.
Amazon followed with $76.95 million, Tesla with $54.08 million, and NVIDIA with $47.04 million in net purchases.
However, their purchase volumes fell well short of one-tenth of SOXL's.
SGOV (iShares 0-3 Month Treasury Bond ETF), a product investing in short-term U.S. Treasury bonds with a maturity of three months or less, which ranked first in net purchases through September 11, continued its net buying streak by purchasing $44.97 million worth.
On the other hand, investors net-sold $69.03 million of SOXS, which tracks three times the inverse daily performance of the Philadelphia Semiconductor Index, opposing SOXL.
Alphabet and Apple were also offloaded, with $24.34 million and $18.05 million sold, respectively.
Although SOXL plummeted to $101.13 on September 14 after reaching as high as $300.77 on June 22, Seohak ants' affection for SOXL shows no signs of cooling down.
This is analyzed to reflect expectations that semiconductors will continue to lead the U.S. stock market despite concerns over overheating investments in artificial intelligence (AI).
In addition, unlike the sluggish domestic market, the U.S. stock market is on an upward trajectory, and the exchange rate briefly dropping to the 1,330 won range for the first time in 1 year and 11 months are cited as factors driving Seohak ants to increase their investments in the U.S. market.
In the domestic ETF (exchange-traded fund) market, funds flocked to "parking-type" ETFs this week, following last week.
Parking-type ETFs are products where cash is placed in short-term financial instruments to earn interest, much like temporarily parking a car.
However, the specific parking-type ETF products that attracted funds changed.
Last week, 552.3 billion won flowed into TIGER Money Market Active, 363.3 billion won into RISE Money Market Active, and 312.5 billion won into KODEX CD Rate Active (Synthetic).
Money market active funds are representative parking-type products that invest in short-term bonds and short-term financial instruments, while CD rate active ETFs are cash-waiting products that track the yield of an index based on 91-day certificate of deposit (CD) rates issued by banks to raise funds.
This week, 830.2 billion won flocked solely to KODEX Money Market Active, while 369.8 billion won and 159.3 billion won flowed into TIGER CD 1-Year Rate Active (Synthetic) and ACD Money Market Active, respectively.
Over the past week, retail investors bought the most of KODEX Leverage (256.7 billion won), while also net-purchasing over 100 billion won each in TIGER Dividend Covered Call Active (109.4 billion won) and KODEX 200 Target Weekly Covered Call (101.6 billion won).
※
Copying, redistribution, and unauthorized use in AI training are strictly prohibited.