Invest1 distinct publisher3 min readUpdated
Ratification votes close this week on a deal that pays most of profit sharing in shares. Guaranteed first-year cash drops from 80% to 40%, with a one-time opt-out for the 2026 bonus.
The Investor · Invest desk

Compiled by The InvestorSomething wrong?How this is made
The pool being split is itself a function of the cycle. Last year's agreement removed the cap on profit sharing and funded it with 10% of operating profit, a structure both sides said would hold for ten years [11]. Paying most of that pool in stock, with shares equal to 40% of the total sellable in the year of payment and a further 20% released at 10% a year over the two years after [5], attaches the memory market to the same money twice: operating profit sets how large the pool is, and the share price sets what it is worth on the day it lands, and again in each of the two deferral years.
For anyone who takes the election on the 2026 bonus, year one is not a pay cut in cash terms. Cash comes to 80% of the pool [6], the same proportion last year's structure paid up front [11], a net change of nothing [16]. What moves is the default. Guaranteed cash falls from 80% to 40% of the pool, a gap of 40 points [15]; the deferral stays at two years and 10% a year, and only its currency changes from cash to shares [18]. The election is described as an exception for the 2026 payment [6], so the new structure bites on the bonus after this one.
The share count will be struck at the lowest of three closing prices: the earnings announcement, the cash payment date and the share payment date [10]. SK hynix presents this as stopping a rising price from shrinking the number of shares employees receive [10]. Read from the register's side, the same rule means the company parts with the most shares when the price is weakest, which tends to be when the operating profit funding the pool has already fallen. The source does not say whether those shares come from new issuance or from treasury stock, which is the difference between dilution and cash already spent, and it gives no figure for how much stock the bonus will absorb against the 40 trillion won SK hynix has proposed to buy and retire [9].
That leaves the vote. Members rejected tentative agreements in both 2023 and 2024 [4], a rejection sends that union back to the bargaining table [3], and some members are objecting that a bonus structure agreed for ten years has been rewritten in its first [12], nine years early [17]. Against that, management has loaded in items that pay out now: more in-house welfare points, congratulatory and condolence payments and funeral support, and an in-house mortgage of up to 200 million won extended from employees with multiple children to all married staff [13]. Relief on one side of the ledger, a pay component priced later on the other.
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
For the 2026 PS to be paid early next year, the deal exceptionally lets employees who wish receive the 40% share portion that is sellable that year as cash instead, so employees can receive up to 80% in cash in the first year.
SK hynix and its unions reached a tentative 2026 wage and collective bargaining agreement after roughly two months of talks, centered on paying 60% of performance bonuses in company shares.
Production-line unions at SK hynix's Icheon and Cheongju plants hold ratification votes on the 2026 tentative agreement from the 24th to the 25th, and the union for technical and office workers is expected to complete voting early in the week.
If either union rejects the agreement, that union must return to the bargaining table with management.
SK hynix unions rejected tentative agreements in ratification votes in both 2023 and 2024, leading to renegotiations.
The core of the agreement is to pay 40% of the profit-sharing bonus in cash and 60% in company shares: shares equal to 40% of the total PS can be sold in the year they are paid, and the remaining 20% in shares is paid out at 10% per year over the following two years.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Detailed but single-sourced and largely unattributed
The cluster rests on one publisher. Deal terms are specific and internally consistent (percentages, vesting schedule, price-selection rule, wage comparison, welfare items), which raises credibility above thin-rumour level. But the vote schedule and member sentiment are sourced to unnamed 'industry sources', no union or company official is quoted, no filing or union bulletin is cited, and there is no second outlet to corroborate. Comparative derivations against last year's structure are sound because both structures are described in the same piece.
Signed but unratified; outcome unresolved
Nothing has taken effect. The agreement is tentative and conditional on ratification votes that were still open at publication, and the same unions rejected tentative agreements in 2023 and 2024, so base rates argue against treating passage as done. The 40 trillion won retirement and FCF commitment is also described as a proposal. Partial credit reflects that management and unions concluded talks without a strike and jointly built supplementary measures, and that the first-year cash election gives members a concrete off-ramp.
Passage framed as probable ahead of the evidence
Modestly overstated. The source asserts the deal is 'seen as likely to pass' and a 'realistic compromise', supported only by unnamed industry sources and analysts, while its own reporting notes two consecutive prior rejections, live member dissatisfaction about a one-year-old system being rewritten, and a new ~3,000-member union yet to take a position. The company's 'employees and shareholders grow together' framing is carried without any quantification of shares issued to payroll or dilution relative to the retirement plan. Terms themselves are reported precisely, which keeps the gap small rather than large.
Company and deal-proponent framing dominate the record
The visible incentive structure leans one way: SK hynix benefits from a share-settled bonus that conserves cash while a large buyback-and-retirement proposal is pending, and from the narrative that employees and shareholders grow together; the reporting relays that framing plus unnamed industry sources and analysts who read the deal favourably. Union-side voices appear only as diffuse 'dissatisfaction'. This is an observation about who is speaking in the record, not about accuracy.
Terms credible, outcome and market impact open
Moderate. Confidence in the described mechanics is reasonably high given the granularity of the terms and the fact that both the old and new structures are laid out side by side, making the derived comparisons robust. Confidence in the story's forward claims is low: single publisher, no named attribution, unratified votes with a two-year record of failure, and no quantification of shares issued to payroll or of any dilution offset against the retirement proposal.
invest
Samsung's $72B payout and SK Hynix's $28.7B buyback bet the memory cycle holds1 distinct publisher
invest
Samsung's one-year payout pool is set to outrun its entire three-year program1 distinct publisher
invest
Korea's chip trade now has 11.4 million shareholders, and a very loud downside1 distinct publisher
invest
Korea's leverage curbs moved the chip bet into margin loans rather than out of it1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
en.sedaily.com
1 article · August 22, 2026