On October 8th, the last payment from my house sale cleared.
I should be careful with the word “windfall.” I sold that house for about 40% less than I paid for it. That is where this series started, and nothing since has made it a better deal. What landed in October was not a profit. It was the leftover — what remained after the loan was settled and the paperwork closed.
Still, it was the largest single transfer I have ever seen in my account: roughly $76,700, at roughly ₩1,340 to the dollar.
Within a few hours I moved about $9,700 of it into my ISA, filling the last of this year’s contribution room.
Then I stopped. Nothing else has moved since. About $67,000 is sitting in a parking account — a Korean high-yield savings account that pays a little interest and asks no questions.
That is not what the plan said.
The plan was not vague
I spent most of the summer on it. Not on picking funds — on picking order. In Korea, that turns out to be the part that matters more. Part 3 was the first version I wrote down.
If you invest from the United States, the shape of the question is familiar: Roth or Traditional, 401(k) or brokerage, and the same index fund inside any of them. Korea has its own version of that question, with different boxes and different rules, and almost nothing written about it in English. So here is the map I was working from.
| Container | Korean name | Rough US analogue |
|---|---|---|
| Tax-advantaged wrapper | ISA | Taxable, with an allowance |
| Personal pension | Yeongeum-jeochuk | Traditional IRA |
| Workplace pension | IRP | 401(k) |
| Taxable brokerage | Ilban-gyejwa | Brokerage |
The analogues are rough, and the details are where the money actually is:
ISA. Annual contribution room is ₩20M, lifetime room ₩100M, with a three-year minimum holding period. Gains up to ₩2M are tax-free (₩2.5M for lower-income filers), and whatever is left is taxed at a flat 9.9% rather than being added to your other income. Unused annual room carries forward. This is the best container I have access to, and it is also the one with a hard annual deadline: room that is not used by December 31 does not come back.
Pension accounts. Here is the detail that surprises almost everyone, including me: the contribution limit and the tax-deduction limit are different numbers. You may contribute up to ₩18M a year across pension accounts, but only ₩9M of that earns a tax deduction. Which means there is an entire category of Korean investor deliberately contributing past the deduction — not for this year’s tax break, but for decades of tax-deferred compounding inside the wrapper. I am one of them.
Taxable account. This is where it gets genuinely strange for an American reader. A foreign index ETF listed on the Korean exchange is taxed at 15.4% on both distributions and price gains — and both are classified as dividend income, which means they stack onto your comprehensive income tax. A US-listed ETF bought directly is taxed instead at 22% on capital gains, separately from your other income, with the first ₩2.5M of gains each year exempt. Same index, same exposure, two completely different tax treatments depending on which exchange you bought it on.
Worth doing the arithmetic on that last one, because the answer is not the one most people assume. Say you realize a $7,500 gain in a year. Through a Korea-listed foreign ETF, you pay a flat 15.4% — about $1,155 — with no exemption. Through a US-listed ETF, the first ₩2.5M (roughly $1,860) is exempt and the rest is taxed at 22%: about $1,241. The “cheaper” 15.4% route costs more.
Flip it around and the crossover sits near $6,200 of annual gains. Below that, the US-listed route wins because of the exemption. Above it, the flat rate wins on its face — except that those gains are dividend income, so they pile onto your comprehensive income, and once total financial income crosses ₩20M a year the effective rate stops being 15.4% at all.
Which is a long way of saying: in Korea the same index fund has two different tax bills, the gap between them is small at ordinary sizes, and it flips direction depending on how much you make. I have seen this explained confidently in both directions online. Neither version is wrong; they are describing different gain sizes.
So the plan was never “what should I buy.” It was “which box, in which order, over the next four years.”
The one rule I was confident about
If the money is only passing through, don’t put market risk in it.
The arithmetic is simple. The cash waiting for future ISA room sits in the taxable account for an average of a bit over a year before it moves. And over a one-year holding period, the S&P 500 has finished lower roughly a quarter to a third of the time, historically. That is not a reason to avoid the index — it is the price of owning it. But it is a very good reason not to own it with money that has a departure date.
Because if that money is down when the transfer window opens, I don’t get to wait. I have to sell at a loss to move it.
And the destination account buys the same index anyway. So the risk gets taken either way. The only question is whether I take it in the container that taxes it at 15.4% and stacks it onto my income, or in the one that shelters it.
The sentence I wrote in my notes back in August: one of my accounts is not an investment account, it’s a waiting room.
I still think that is right. It is the most useful thing I learned all year.
Then two things happened
The law moved. In early August, the government proposed capping ISA contracts at five years, scrapping the carry-forward of unused room, and sunsetting the whole thing in 2029. I built part of my plan around dodging that — my account happens to have an unusually long maturity date, set years ago for no clever reason at all.
On September 1, the cabinet confirmed the final bill and all three of those changes were gone. Contract length, contribution limits, and carry-forward all stay as they are. It still has to pass the National Assembly in December, so nothing is final. But the cleverest-looking part of my plan — the part where my old account quietly escaped a new restriction — turned out to be irrelevant. It wasn’t foresight. The restriction simply never happened.
And I moved. Thirty-year Treasury yields touched their highest level since 2006 last month. The S&P 500 closed above 7,800 for the first time in early October, and the Nasdaq set records in the same stretch. Long rates screaming one thing, equity indexes screaming the opposite, at the same time.
I looked at that and did not want to convert that $67,000 into anything.
I want to be accurate about what that is. It is market timing. I have spent two years writing that I don’t do market timing, and the honest version of this post is that when a number with eight digits landed in my account, I did some.
What I can defend, and what I can’t
Defensible: the automatic monthly purchase never stopped. About $1,137 a month goes into index funds across my pension accounts on schedule, the way it has since 2024. (I started investing in 2009; the index part is only two years old, and the seventeen years before it were mostly an expensive education.) None of that paused. The system that actually built the portfolio is still running, untouched by any of this.
Not defensible: holding cash is not free. If the index rises 10% over the next year, that hesitation cost me something real, and no amount of careful reasoning afterward will refund it. Safety has a price. I have written that sentence before, about other people.
There is also a distinction I had to be walked through, and it is worth passing on. The waiting-room holding I planned to use — an ultra-short Treasury fund — is not really an investment. I classify it as cash in my own portfolio tracking, because that is what it is: a way to hold dollars. Which means “I’m delaying because the market scares me” doesn’t actually describe what I’m delaying. Moving cash into a cash-equivalent carries almost no market risk. What I would actually be deciding is when to convert won into dollars, and whether I want that exposure now.
Those are two different questions. I had them fused into one, and fused they were impossible to answer.
If you are somewhere similar
Not advice — just the things I would tell myself in August.
Know your containers before the money arrives. The week the funds land is the worst week to start reading about contribution limits. The reading is slow; the deadlines are not.
Separate the deadlines from the decisions. Annual room that expires on December 31 is the part you genuinely cannot redo. I filled mine in a few hours, and that is the one piece of the plan I executed perfectly. Everything else can wait a month without costing anything.
Separate where it sits from when you buy. Parking money in a cash account and buying an index fund are two decisions. Deciding them together means you will postpone both.
A plan that assumes the tax code holds still is a fragile plan. Mine assumed four years of stability and got a proposed overhaul within four weeks, followed by a reversal within four more.
Write down the reason, not just the decision. I can tell you exactly what my plan says. I am having a harder time explaining why I stopped following it, and that difficulty is itself the useful signal.
Where this actually stands
About $67,000 is in a savings account and I have not decided what to do with it. Roughly $44,000 of it has a designated path — it funds the next three years of ISA contributions, one tranche each January. The remaining $23,000 has no assigned home at all; it was earmarked for something I have since set aside.
I am not going to pretend I have landed somewhere. I will report the number in next month’s recap either way, moved or not, which is the only promise this blog has ever really made.
This is not investment advice. I’m an engineer writing down what I actually did with my own money, mistakes included. Tax rules described here are Korean and were accurate as of October 2026; the 2026 reform bill is still before the National Assembly.
