Invest1 distinct publisher3 min readPublished
The 167 per cent jump disclosed to the Financial Supervisory Service is a March settlement on last year's books, which makes it hard evidence about the previous leg of the memory cycle and leaves next March to test this one.
The Investor · Invest desk
Compiled by The InvestorSomething wrong?How this is made
Tax is a lagging indicator by design, and Korea's design lags by roughly a year. Corporate tax here is finalised and paid in March on the prior year's results, with an interim payment in August calculated on provisional first-half earnings [6], so the 11.21 trillion won that left Samsung Electronics and SK hynix in the first half [1] settles profits booked before any of this year's numbers existed, while the operating profit everyone is actually watching (about 146.7 trillion won at Samsung and 98.2 trillion at SK hynix in the first half, six to 13 times a year earlier [7]) has not been taxed at all yet. Which is why dividing tax paid by profit earned over the same six months yields 2.7 per cent at Samsung and 7.4 per cent at SK hynix [11], two fractions that describe a calendar rather than a rate.
The comparison that does the analytical work is the trough. The 4.19 trillion won paid in the first half of last year [2] was a third of the 12.66 trillion won paid in the first half of 2019 [5], and even after nearly tripling, this year's total is still 1.45 trillion won short of that 2019 mark [4]. SK hynix supplied 64 per cent of the combined figure [2] and set its own half-year record, 59 per cent above its 2019 high [3]. So the honest reading of the print is that the upswing is cash and not guidance, or rather that last year's upswing was, which is a smaller claim than the headline number invites and a more useful one.
The talked-about 100 trillion won annual figure [10] is more conservative than its own inputs, which is worth showing. Consensus full-year operating profit is 385 trillion won at Samsung and 266 trillion at SK hynix [8], 651 trillion combined [6], and 20 per cent of that is 130 trillion [7]; the projection instead applies the 20 per cent rate to a standalone base of 500 to 600 trillion [9], a haircut of between 8 and 23 per cent for the gap between consolidated and standalone books [8]. Against the 15.64 trillion won combined record of 2019 [9], even the low end is about 6.4 times [10].
Everything then hangs on the 20 per cent, where each percentage point of effective rate is worth 5.5 trillion won on the midpoint of that base [9]. Both companies are expanding facility investment and research spending, and the credits attached to that spending can pull the realised payment below the simple estimate [11], which means the state's take next March is partly a function of how much fab and lab money the two firms decline to spend, and neither shows much sign of declining. This is probably wrong, but I would take the under: a credit-heavy year of construction reads as a seven or an eight at the front of the March number rather than a one and two zeros. The counter-thesis is respectable and may well win, and an industry official's framing that better chip earnings feed tax revenue and state finances [12] is directionally correct; if standalone profit lands near 600 trillion and credits stay modest, the settlement clears 120 trillion and the line re-bases for at least a year. What would falsify my caution is disclosed credit take-up small enough to keep the realised rate above about 17 per cent. What would confirm it is a 2019-style sequel, given that this same line item fell by two-thirds between booms [5].
Ranked by verification strength, evidence, and original report placement.
First-half operating profit reached about 146.7 trillion won at Samsung Electronics and 98.2 trillion won at SK hynix, six to 13 times higher than a year earlier.
Samsung Electronics paid 3.99 trillion won and SK hynix paid 7.22 trillion won in corporate tax in the first half, a combined 11.21 trillion won, according to the Financial Supervisory Service on the 30th.
The combined 11.21 trillion won marks a 167% jump from 4.19 trillion won a year earlier.
The combined figure is the second-highest half-year total ever, trailing only the 12.66 trillion won paid in the first half of 2019 during the previous chip boom.
Samsung's corporate tax payments rose 256% from a year earlier and SK hynix's climbed 135%.
SK hynix set a company record for a half-year period, surpassing its previous high of 4.53 trillion won in the first half of 2019.
Distinct publishers with included, body-backed reporting in this cluster.
en.sedaily.com
1 article · August 30, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
invest
Korea's chip trade now has 11.4 million shareholders, and a very loud downside1 distinct publisher
invest
IIF pins the won's July break from the yen on Samsung and SK hynix fab spending1 distinct publisher
invest
Muan County takes its seat, and an 800 trillion won fab cluster gets a committee date1 distinct publisher
invest
Korea's leverage curbs moved the chip bet into margin loans rather than out of it1 distinct publisher
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.
Regulator-sourced core, single set of eyes
The two figures that carry the story — 3.99 trillion and 7.22 trillion won — come out of Financial Supervisory Service filings, and en.sedaily.com states the March-settlement rule that decides what a 167% jump actually means. Against that: one outlet, one disclosure date, and profit lines printed beside the tax that work out to 2.7% and 7.4% ratios, a mismatch that resolves only because the cash and the profit belong to different years. Nobody independently re-derives any of it.
No demand data, only its cash residue
Tax paid is what demand leaves behind a year later, and our coverage contains nothing else: no shipment volumes, no contract pricing, no named customers, no capacity commitments. The one hard disclosure describes the previous leg of the memory cycle, and the only forward-looking figure is an analyst estimate. There is no basis to score how widely the current boom is landing.
Headline projection outruns the disclosure
The verified event is a large tax payment on the cycle that already happened; the headline is a 100 trillion won bill that has not been assessed yet. That projection needs a profit base trimmed to somewhere between 77% and 92% of consensus, a clean 20% effective rate, and no meaningful credit offset — and en.sedaily.com itself flags the last of those as a reason payments could come in lower. A 6.4-fold jump over the 2019 record is a big claim to hang on two assumptions, and next March is the only thing that settles it.
Taxpayer-of-the-year framing, credit-seeking industry
Only one voice interprets the numbers, and it is an unnamed industry official arguing that chip profits are good for the treasury. That argument is useful to a sector the same report describes as expanding the facility and R&D spending whose credits could shrink the bill — being seen as the state's biggest taxpayer is helpful when the ask is more relief. Not disqualifying, but the interpretation and the interest run in the same direction.
Solid on the past, speculative on the present
Split the story and the confidence splits with it. What was paid is documented and dated; what will be paid is an extrapolation from one outlet with an acknowledged offset and no second source to test the profit assumptions. Treat the 11.21 trillion won as hard, the 100 trillion won as a scenario, and revisit in March.