Stocks
US retail is mildly bearish on US stocks. Korean and Japanese retail are buying them anyway. US retail is mildly bearish on US stocks. Korean and Japanese retail are buying them anyway.
The AAII sentiment survey for the week ending August 12, 2026 came in at 34.7% bullish, 27.4% neutral, 37.9% bearish. That is a bull-bear spread of -3.2 points. Long-run averages are 37.5 / 31.0 / 31.5, so bullishness sits about 2.8 points below normal and bearishness about 6.4 points above it. Nothing dramatic. Mildly gloomy, close to the middle of the range.
Here is the same market seen from two other places.
**Korea**
Korean retail investors net bought $4.642 billion of US equities between July 3 and August 3. The monthly pattern this year ran $5.003B in January, $3.949B in February, $1.692B in March, then net selling of $469M in April and $940M in May, back to $633M of buying in June, then July. Total custody of US stocks passed 300 trillion won, an all-time high.
The composition is the part worth stopping on. The single most bought security over that month was Direxion Daily Semiconductors Bull 3x at $3.695 billion — roughly four fifths of the entire net purchase sitting in one leveraged semiconductor ETF. Second was SK Hynix ADR at $868.7M, third a 3x QQQ product at $399.9M, then Alphabet and SpaceX.
**Japan**
A survey of 800 Japanese individual investors published July 1 found 67.3% expecting the Nikkei to rise further in the second half. The most common year-end call was 75,000 yen, and 39.4% named 80,000 or higher. Investors in their twenties were the most bullish cohort at 58.9%, median call 89,000.
**What this is and is not**
These are three different questions. AAII asks about direction over six months. The Japanese survey asks for a year-end index level. The Korean figure is money that actually moved, not an opinion at all. You cannot put them on one axis, and I would not trust anyone who did.
What survives the caveats is narrower. At a moment when American retail sentiment is running a little below its own average on its own market, foreign retail money is arriving in size, and arriving leveraged. Whether that reads as contrarian, late-cycle, or noise is the argument, and I do not have a view worth much.
If you watch US retail flows, does any of this show up on your side?The AAII sentiment survey for the week ending August 12, 2026 came in at 34.7% bullish, 27.4% neutral, 37.9% bearish. That is a bull-bear spread of -3.2 points. Long-run averages are 37.5 / 31.0 / 31.5, so bullishness sits about 2.8 points below normal and bearishness about 6.4 points above it. Nothing dramatic. Mildly gloomy, close to the middle of the range.
Here is the same market seen from two other places.
**Korea**
Korean retail investors net bought $4.642 billion of US equities between July 3 and August 3. The monthly pattern this year ran $5.003B in January, $3.949B in February, $1.692B in March, then net selling of $469M in April and $940M in May, back to $633M of buying in June, then July. Total custody of US stocks passed 300 trillion won, an all-time high.
The composition is the part worth stopping on. The single most bought security over that month was Direxion Daily Semiconductors Bull 3x at $3.695 billion — roughly four fifths of the entire net purchase sitting in one leveraged semiconductor ETF. Second was SK Hynix ADR at $868.7M, third a 3x QQQ product at $399.9M, then Alphabet and SpaceX.
**Japan**
A survey of 800 Japanese individual investors published July 1 found 67.3% expecting the Nikkei to rise further in the second half. The most common year-end call was 75,000 yen, and 39.4% named 80,000 or higher. Investors in their twenties were the most bullish cohort at 58.9%, median call 89,000.
**What this is and is not**
These are three different questions. AAII asks about direction over six months. The Japanese survey asks for a year-end index level. The Korean figure is money that actually moved, not an opinion at all. You cannot put them on one axis, and I would not trust anyone who did.
What survives the caveats is narrower. At a moment when American retail sentiment is running a little below its own average on its own market, foreign retail money is arriving in size, and arriving leveraged. Whether that reads as contrarian, late-cycle, or noise is the argument, and I do not have a view worth much.
If you watch US retail flows, does any of this show up on your side?
0
Comments 0