Kim Ki-wan, a private banker at Korea Investment & Securities, won the 2025 Hankyung Star Wars competition with a 53% return but underperformed the KOSPI index in the first half of this year. His top performer was SK Hynix. Kim attributed the underperformance to misjudging large-cap rally potential and overestimating small-cap rebound timing. The Korean stock market experienced extreme semiconductor concentration, with the advance-decline ratio falling to historic lows.
Kim Ki-wan Admits Large-Cap Rally Misjudgment
Kim stated that his biggest mistake was underestimating how much large-cap stocks could rise. He said large-cap stocks like Samsung Electronics, SK Hynix, and Samsung Electro-Mechanics are often assumed to have limited upside, but when strong trends form, they can deliver unexpected gains and continue rising. He emphasized that investors should not arbitrarily set limits on large-cap potential.
For small-cap stocks, Kim noted that he made the naive assumption that stocks would rebound after falling a certain amount. He warned that stocks can fall further than expected and that investors should develop specific scenarios for when and why a rebound will occur, rather than relying on the assumption that "it has fallen a lot, so it will bounce back."
Portfolio Allocation: 50% Semiconductors for Second Half
Kim plans to maintain a 30-40% allocation to large-cap semiconductor stocks like Samsung Electronics and SK Hynix in the second half. He will allocate an additional 10-20% to semiconductor materials, parts, and equipment suppliers that benefit from investments by integrated device manufacturers (IDMs). His total semiconductor sector allocation will be around 50%.
He will allocate 20-30% to power equipment and renewable energy sectors related to AI data centers. Kim noted that many companies with strong growth prospects and market positions experienced larger-than-expected corrections due to semiconductor supply concentration. He believes buying these companies at appropriate prices could yield high returns even if held only until previous highs.
The remaining 20% will be allocated to cash and undervalued sectors. Kim highlighted that the ADR has fallen too low, creating potential for rebounds in other sectors. He plans to invest around 20% in deeply undervalued sectors like consumer goods and biotech. He noted that biotech fundamentals have improved significantly compared to the past, creating opportunities to buy quality stocks cheaply.
Stock Selection Method: Top-Down Approach
Kim stated that he traditionally invested based on individual company fundamentals, evaluating companies by their price-to-earnings ratio (PER) relative to future earnings growth. However, he now believes that in the current market, a top-down approach — analyzing macroeconomic trends and sectors first, then selecting sector leaders — is more important for maximizing returns. He is adapting his method to incorporate more top-down analysis.
For consumer goods investments, Kim said he gains investment ideas by using products directly and visiting sales channels like department stores and Olive Young. He also tracks trend data that others do not continuously monitor. He noted that for sectors with less market attention, data may already show positive trends that the market has not yet reflected.
Kim cited his investment in Paradise, a casino and hotel company, as an example. Foreign visitor numbers and monthly casino performance data were improving, but even sector analysts were not incorporating this into their analysis. He identified this in advance and invested, profiting when the stock rose on earnings surprises.
Renewable Energy Sector Undervalued in AI Value Chain
Kim stated that the renewable energy sector, including wind and solar, has not yet been fully valued within the AI value chain. Nuclear power and power equipment sectors saw stock price increases early in the AI trend, but renewable energy did not receive immediate stock price reflection due to the perception that intermittency issues would limit its use in AI data centers.
However, Kim noted that recent trends show energy sources are insufficient overall, so all available energy — nuclear, solar, wind — is needed. He said that energy storage systems (ESS) can mitigate the intermittency weakness, making renewables viable. While the sector gained attention during the US-Iran war due to oil price spikes, Kim believes long-term recognition of renewables' role in the AI trend is still lacking, warranting increased attention.
Selling Strategy: Scenario-Based Exits and -15% Stop-Loss
Kim said he develops selling scenarios for each stock at the time of purchase. For companies he expects to grow long-term, he continues monitoring and does not sell arbitrarily unless the investment thesis is proven wrong. If growth prospects and the story remain valid but the price declines, he sometimes adds to positions. However, if the thesis is wrong and long-term growth appears unlikely, he sells immediately regardless of current price.
For stocks purchased with a trading perspective, Kim records stop-loss levels and strives to adhere to them. He emphasized that maintaining stop-loss discipline is critical for risk management, especially when chasing stocks during periods of strong market attention. His average stop-loss level is around -15%, though it varies by stock price movement. He stated that selling timing is determined by the concept under which the stock was purchased.
FAQ
What was Kim Ki-wan's biggest investment mistake in the first half of this year?
Kim stated his biggest mistake was underestimating how much large-cap stocks like Samsung Electronics and SK Hynix could rise when strong trends form. He also overestimated rebound timing for small-cap stocks, assuming they would bounce back after falling a certain amount without developing specific scenarios for when and why rebounds would occur.
How will Kim Ki-wan allocate his portfolio in the second half?
Kim plans to allocate around 50% to semiconductors (30-40% to large-cap stocks like Samsung Electronics and SK Hynix, 10-20% to materials/parts/equipment suppliers), 20-30% to power equipment and renewable energy sectors related to AI data centers, and the remaining 20% to cash and undervalued sectors like consumer goods and biotech.
Why does Kim Ki-wan believe renewable energy is undervalued in the AI value chain?
Kim noted that renewable energy sectors like wind and solar have not been fully valued within the AI value chain due to intermittency concerns. However, he stated that energy sources are insufficient overall, so all available energy is needed. Energy storage systems can mitigate intermittency, making renewables viable for AI data centers. Long-term recognition of renewables' role in the AI trend is still lacking.
