I woke up Friday morning, checked my phone before coffee like I always do, and saw a headline that made me sit up: the Kosdaq had tripped a sidecar (사이드카) — a five-minute halt on program sell orders — and closed down 4.63%. Meanwhile the Kospi, sitting right next to it, actually finished the day up almost a full percent. Same country, same morning, two completely different outcomes. That gap is the whole story.
I don’t own a single Kosdaq stock. My accounts are built around S&P 500 and Nasdaq 100 index funds, a dividend-focused Dow Jones fund, and a slice of long-dated U.S. Treasury bonds. So my first reaction to the sidecar headline wasn’t panic — it was curiosity. Why did the growth-heavy Kosdaq get hit so much harder than the blue-chip-heavy Kospi on the same day, from the same set of headlines?
The answer turned out to be a concept I’d only ever applied to my bond fund before: duration.
What actually happened Friday
Korean exchange data show the Kosdaq closed at 801.94, down 38.95 points, or 4.63%, on August 21. A sell-side sidecar triggered around 10:05 a.m. after Kosdaq 150 futures fell more than 6% and the cash index dropped over 5%. Reporting on the selloff points to two forces feeding it together: a jump in U.S. long-term Treasury yields and a jump in oil prices, layered on top of profit-taking after the prior day’s gains. Battery, biotech, and robotics names — the more speculative, growth-oriented corners of the Kosdaq — led the decline.
That “long-term yields” piece isn’t new information to me. The 30-year U.S. Treasury yield touched roughly 5.3% around August 17–18, a 19-year high, before Wednesday’s Treasury debt-buyback announcement briefly pulled it back down — then it rebounded to about 5.25% by Thursday’s close as investors decided the relief wouldn’t last. I wrote about what rising long-term yields do to a bond fund a couple of weeks ago, after my own 30-year Treasury holding took a hit. What I hadn’t connected until Friday is that the exact same mechanism — a rising discount rate — also explains why growth stocks got hurt worse than steadier stocks on the same day, in the same market.
Oil was the other half of the story. Brent crude closed at $93.78 a barrel on August 20, its highest level since late July, after the U.S. Treasury Secretary said Washington was preparing the “most severe” economic sanctions ever on Iran. Walmart’s earnings that same week were solid on paper — the company beat forecasts — but management flagged that gas prices were pushing shoppers toward trade-offs, spending less on discretionary items to cover fuel. Rising oil doesn’t just squeeze gas budgets; it feeds into the same inflation and rate expectations that pushed Treasury yields to a 19-year high in the first place. U.S. stocks felt some of this too — the Dow fell about 1.3%, the S&P 500 about 0.8%, and the Nasdaq about 1% on Thursday, before the Kosdaq’s sharper move on Friday.
Duration isn’t just a bond word
Here’s the mental shortcut I’d been missing. A bond’s price sensitivity to interest rates — its duration — comes from a simple fact: a dollar you’re promised further in the future is worth less today when rates rise, and it’s worth a lot less if that dollar is a long way off. A 30-year bond’s payments are almost entirely far in the future, so its price swings hard when rates move. A short-term bond’s payments arrive soon, so it barely moves.
Stocks work on the same math, even though nobody hands you a fixed coupon. When investors price a stock, they’re implicitly discounting all its future earnings back to today’s dollars — the same rate-and-currency machinery I mapped out a few weeks ago. A mature, cash-generating company — the kind trading at a modest earnings multiple — is priced mostly on what it’s expected to earn over the next few years. A younger, fast-growing company, especially one trading at a rich multiple because the market is betting on earnings much further down the road, is priced on cash flows that are, in effect, further away in time. In interest-rate terms, that growth stock behaves like it has a longer duration than the mature one, even though neither of them is a bond.
That’s why a market like the Kosdaq — tilted toward battery, biotech, and robotics names still years from their biggest expected profits — reacts more violently to a spike in long-term rates than a market like the Kospi, which is anchored by semiconductor and industrial giants already generating today’s earnings.
Doing the arithmetic, not the forecasting
I wanted to see the size of this effect in real numbers, so I ran the present-value math instead of guessing.
Take a single dollar of profit a company is expected to earn two years from now, and another dollar expected ten years from now. Discount both at 5%, then discount both again at 6% — roughly the move in long-term Treasury yields we’ve just watched happen.
At a 5% discount rate, the two-year dollar is worth about 90.7 cents today; the ten-year dollar is worth about 61.4 cents. Bump the discount rate to 6%, and the two-year dollar falls to about 89.0 cents — a loss of roughly 1.9% of its value. The ten-year dollar falls to about 55.8 cents — a loss of roughly 9.0% of its value.
Same one-percentage-point rate move. The near-term dollar lost about 1.9% of its present value; the far-off dollar lost about 9.0% — nearly five times more. That gap, multiplied across a whole company’s future earnings, is a rough sketch of why a richly-valued growth stock can fall several times harder than a steady earner when long-term rates jump, even with no change in either company’s actual business. This is arithmetic, not a forecast — real stock prices depend on far more than a single discount rate, but the direction and rough scale of the effect are real.
A structural comparison: sidecar vs. circuit breaker
Korea and the U.S. both have mechanisms to slow a market down when it’s falling fast, but they’re built differently.
| Korea’s sidecar | U.S. circuit breaker | |
|---|---|---|
| What halts | Sell program orders | Entire market |
| Typical trigger | Futures move ~6% | Index down 7/13/20% |
| Duration | Five minutes | 15 min to close |
| How often | Several times a year | Very rare |
A Korean sidecar is a narrower tool — it pauses computer-driven sell programs for five minutes so human judgment can catch up, without stopping the whole market. A U.S. market-wide circuit breaker is a blunter instrument that halts everything, and it’s calibrated to trigger only in genuinely rare, severe selloffs. Neither mechanism changes the underlying math above; they just buy the market a few minutes to absorb it less chaotically.
Where my own accounts sit in this picture
I don’t hold individual growth names, Korean or American. My retirement-advantaged accounts — two personal pension savings accounts (연금저축), an IRP, or individual retirement pension (개인형퇴직연금), and an ISA, or Individual Savings Account (개인종합자산관리계좌) — are built around S&P 500 and Nasdaq 100 index funds, a dividend-tilted Dow Jones fund, a 30-year Treasury bond fund, and cash reserves. That structure means I never have to guess which single battery or robotics company will be the winner. But it doesn’t make me duration-proof. The Nasdaq 100 in particular is packed with the same kind of long-duration, high-multiple companies that make up the riskier end of the Kosdaq — just American ones. Owning an index instead of a single stock diversifies away company-specific risk. It does not diversify away interest-rate risk, because if long rates keep climbing, the whole growth-heavy end of any index feels it.
It’s worth pausing on the account structure itself, because Korea and the U.S. solve the same problem — encouraging long-term saving — with different tools.
| Korea | U.S. | |
|---|---|---|
| Personal tax-deferred account | Personal pension savings | Traditional IRA |
| Employer-linked account | Retirement pension | 401(k) |
| Tax-sheltered general account | Individual Savings Account | Roth IRA (rough parallel) |
My personal pension accounts are closest in spirit to a Traditional IRA: individually opened, invested in what I choose, with a tax deduction on contributions up to an annual cap. My IRP works similarly but sits alongside a small employer contribution, which is closer to how a 401(k) blends employee and employer money, though the two systems are structured differently underneath. My ISA doesn’t have a clean U.S. equivalent — it’s a general-purpose, time-limited account that shelters investment gains from tax once you commit to holding it for a few years, sitting somewhere between a taxable brokerage account and a Roth IRA in how it behaves. None of these accounts change what duration does to the assets inside them; they only change how much of the return I get to keep.
Where this reasoning cuts against me
The honest counter-argument is that growth stocks have historically outperformed steadier, value-oriented stocks over long stretches, and that outperformance is supposed to be the reward for exactly the risk I just described — a risk premium for tolerating bigger swings. If I flinched every time long rates spiked and rotated away from growth-heavy index exposure, I could end up giving away decades of higher expected returns to avoid a handful of painful weeks. It’s the same discipline question I ran into when I compared picking individual stocks to just holding the index: the data usually rewards staying put, not reacting. There’s also a real question of how much Friday’s move actually matters to a Nasdaq 100 fund I plan to hold for another twenty years, versus how much it matters to a Kosdaq trader who needs the price to hold up by next Tuesday. My time horizon is long enough that a single rate-driven selloff, however sharp, is designed to be background noise rather than a reason to change anything. The discipline isn’t in predicting these days — it’s in not reacting to them.
A few questions I would have asked myself
Is a sidecar the same thing as a circuit breaker? No. A sidecar is a narrower, shorter pause aimed at computer-driven sell programs; a circuit breaker is a broader halt of the entire market, reserved for much larger, rarer drops.
Why do growth stocks fall more than steadier stocks when rates rise? Because more of a growth stock’s expected value sits in profits further in the future, and future dollars lose more present value than near-term dollars when the discount rate goes up — the same math that makes long-term bonds more rate-sensitive than short-term ones.
If I only own the S&P 500, am I safe from this? Safer than owning individual growth names, but not immune. Large index funds still include plenty of long-duration, high-multiple companies; the effect is smaller and more diversified, not absent.
Does this mean I should sell my growth exposure before rates rise further? That is a market-timing call, and market-timing calls are exactly what dollar-cost averaging into a broad index is designed to avoid making. I am not making one here.
Friday’s sidecar didn’t change anything about my own accounts. I didn’t buy, sell, or touch a single position. But it gave me a clearer picture of a mechanism I’d only half understood — one that connects a Treasury yield chart, an oil price, a Kosdaq halt, and my own Nasdaq 100 fund into a single, coherent story. That is usually the most useful thing a volatile day can hand you: not a trading signal, but a better model of how the machine works.
This is not investment advice.