AVI’s McGarry: Korean opportunities beyond the memory cycle

Several sectors remain undervalued, according to the senior investment analyst

3–4m

By Ross McGarry, senior investment analyst, AVI Global Trust

The South Korean semiconductor industry has historically been a deeply cyclical one. The vicissitudes and resulting mergers and bankruptcies have taken the world from approximately 30 players to just three today: Samsung Electronics, SK Hynix and Micron.

Against this constrained supply, the advent of AI has sent demand parabolic, with a memory super-cycle driven by accelerating High Bandwidth Memory demand, pushing memory prices higher. One metric which illustrates the immense magnitude of this earnings inflection is the upward revisions to Samsung’s 2026 consensus earnings before interest and tax, where forecasts have been revised from 137% year-on-year growth at the start of 2026 to around 700%. On current estimates, Samsung is on track to become the second most profitable company on earth in 2027, behind only NVIDIA.

Goldman Sachs paints an equally eye-catching picture at the index level, with consensus expectations for 2026 earnings revised from 48% at the start of the year to 277%. Even stripping out Samsung Electronics and SK Hynix, the rest of the market has also seen significant upgrades to 2026 profit expectations – a reminder that the earnings upgrade cycle, while concentrated, is broader than many appreciate.

See also: GWA’s Alexander: Reasons to be cheerful

Samsung Electronics and SK Hynix now represent approximately half of South Korea’s KOSPI index. Both are trading at historically elevated valuations compared to previous cycles, reflecting investor confidence that this memory upcycle may prove more durable than previous ones due to structural changes which could support a more sustained earnings recovery. Nonetheless, the industry remains cyclical and even with improved long-term supply contracts, this current super cycle will at some point turn. Calling when this will be is, of course, the billion – or even trillion – dollar question and one the many market participants are trying to figure out.

The market beneath the surface

The KOSPI’s headline price-to-book now stands at 2.5x which is in-line with the emerging market average. This is a remarkable achievement in just over twelve months, having re-rated from 0.9x in December 2024. However, this flatters the breadth of the re-rating, with two thirds of all companies on the KOSPI still trading below book value and 41% below 0.5x. This is significantly higher than in equivalent markets such as Japan, the US or Europe.

The same can be said of the index performance as, year to date, the KOSPI has returned 118%, while the equal weighted KOSPI has returned 49%, a vast difference in performance which has only been widening. The headline index performance obfuscates the abundance of value on offer for bottom-up investors in the Korean market.

The evidence of this is striking. Screening the roughly 2,300 non-financial listed companies and applying both liquidity and capitalisation thresholds leaves an investable universe of over 600 names. Within it, shares trade at an average 0.65 times book value, with assets backing market capitalisation by 130% on average. The market, in other words, is paying markedly less than the underlying assets are worth.

There is also further hidden value on offer in the form of non-core real estate holdings, with numerous companies holding significant portfolios of investment properties on their balance sheets that are notoriously hard to screen for.

See also: Lessons UK can learn from US’s move to T+1 

Yet this undervaluation is not coming at the expense of quality, with the top 100 companies in this universe ranked by quality metrics displaying average operating margins (EBIT) of 17% and returns on capital of 12%. Despite this underlying business quality, these names have generated a median total return of -7% year-to-date in local currency, against the KOSPI’s return of 118%, and have barely participated in the re-rating story since it began last year.

Many of these companies are also at the centre of the government’s focus for structural reform – holding company discounts, treasury share cancellation, excess capital, dual listings, low price-to-book ratios and low payout ratios. All told, despite the hype, the majority of investable stocks in Korea remain unloved, undervalued and under-researched, with the valuations of names not reflecting the underlying business quality, and with significant asset backing which makes them ripe for engagement.

The opportunity in Korean equities is as exciting as it has been at any point in our time following this market and more can – and needs – to be done to sustainably drive a re-rating across all companies.