See, over the past two days, the KOSPI has dropped from 6,000+ to 5,440, falling by nearly 13%. Samsung and SK Hynix both took a brutal hit. At first, I thought it was just a general market issue, but once I looked deeper, I realized the real problem is much more complicated.

Let’s step back a bit. The South Korean market has been surging all through the past year. The KOSPI gained 75.6% in one year—number one in the world. Why did it rise so sharply? If you look closely, about half of those gains came from just two stocks: Samsung and SK Hynix. These two companies produce HBM—high-speed memory chips that NVIDIA needs for AI. In fact, there are only three companies worldwide that can actually produce HBM, and Korea accounts for more than 80% of that market.

So how does energy come into play? This is the interesting part. South Korea has no natural gas or coal of its own; it has to import everything. And in February, Iran began closing the Strait of Hormuz—a crucial passage through which about one-fifth of the world’s oil and a huge amount of natural gas flow. Natural gas from Qatar, Korea’s main source, has to pass through this strait as well. Gas prices surged by nearly 40% across Asia.

Now, do the pieces connect? Samsung and Hynix need a lot of electricity to produce chips, and that electricity comes from natural gas. Natural gas prices have gone up, and on top of that, energy uncertainty has rattled foreign investors. On 27 February, they sold off 6.8 trillion won. Not long after, they sold an additional 5.1 trillion won. In just half a year, half of it simply disappeared.

So who stepped in to take over the job? South Korean retail investors. They saw the price drop to levels not seen in ten years and thought it was an opportunity. Unfortunately, the market kept posting losses. On 3 March, they bought. On 4 March, prices fell again. The market didn’t stabilize just because retail investors entered—foreign investors kept selling continuously.

What I notice from this is that the South Korean market is nicknamed the “Korean Discount.” Because for the same company, if it’s listed in Korea, the price is lower than if it’s listed in the US or Japan. The old problems—family-style management and low dividend payouts—persist. The new president is trying to address this. For a while, a lot of foreign capital flowed in, and the market surged by 75% in a year. It seemed like the discount was gone. But over the past two days, something new has surfaced.

The deeper problem is that the Korean market depends on only two stocks—and those two stocks depend on energy, which Korea can’t control. The whole market ends up betting on a single industry. When something happens outside that industry, the market stalls immediately. It took 34 days to rise from 5,000 to 6,000, but it only took two days to fall from 6,000 to 5,440.

The reality is that demand for AI computing power remains real, and the HBM gap is still real. Big customers still need chips. But these past two days made everyone understand something important: the rise is driven by fundamentals, while the fall is driven by emotions. Fundamentals move slowly; sentiment moves fast. If someone runs fast, they get through first. Those who arrive late have to wait for the Strait of Hormuz to reopen, or wait for liquidity to come into the market from elsewhere.