The number impresses until you look at where it comes from. Of the ninety-six
companies in the index, practically the whole move is explained by two:
Samsung and SK Hynix, the world's two largest memory chip makers,
which together weigh around 46% of the index.
Why that matters and is not a technicality
A cap-weighted index is not an average of a country: it is an average of the
money. When two companies weigh almost half, buying the Korean index is not
diversifying into South Korea — it is buying memory chips with a little of
everything else on top.
It is the same phenomenon that makes the S&P
500 depend on a handful of tech names, only taken to the extreme. The
difference is that in Korea's case the sector is a single, deeply cyclical one.
The part that genuinely worries
The euphoria stopped being a fund manager's affair. There are retirees in South
Korea mortgaging their homes and taking on credit to buy these two stocks, because
they see them rising fast and because other people are making money.
That is the symptom that matters. Not the price:
the disappearance of any sense of risk. When someone pledges
their house to buy shares in a cyclical sector, they are not calculating an
expected return: they are reacting to the fear of missing out.
And there is a second signal in the same place: leveraged
ETFs appeared on each of those two stocks. Those
products do not exist because someone asked politely: they exist when there is
enough demand from people willing to multiply their exposure. They are a
consequence of the euphoria, not a cause.
The catalyst that could correct prices
None of this means Samsung and SK Hynix are not worth what they are worth.
Memory demand for data centres is real and large.
But there is one concrete fact that can change the arithmetic: China's
CXMT began producing high-speed memory at a lower price. In a
commodity business — and memory is one, however much it does not look like it
right now — a new competitor with aggressive pricing is the classic catalyst for a
correction.
The lesson, which is not about Korea
Before buying an index, check what its top three positions weigh. If they add
up to more than a third, you are not buying a country or a market: you are buying
those three companies in a wrapper that looks like
diversification.
Frequently asked questions
Why did South Korea’s Kospi rise so much?
The index rose 27% in May, but of its ninety-six companies almost the entire move comes from two: Samsung and SK Hynix, the world’s two largest memory chip makers, which together carry around 46% of the index.
Is buying the Kospi the same as diversifying into South Korea?
No. A market-cap weighted index is not an average of a country, it is an average of the money. With two companies carrying nearly half the weight, buying the Kospi means buying memory chips with a little of everything else on top — and memory is a deeply cyclical business.
What signs of euphoria showed up in the Korean market?
Retirees mortgaging their homes and borrowing to buy those two stocks, and leveraged ETFs launched on each of them. Neither is a cause: both are consequences of risk perception disappearing. The concrete catalyst to watch is China’s CXMT, which started producing cheaper high-speed memory.
