Hyundai E&C vs Daewoo E&C: Korea’s Nuclear Builders, Decoded — and One of Them Has No Contract (2026)

In July 2024 the headlines said Korea had beaten France to a ₩24 trillion (about $17bn) Czech reactor order. Since then the company named as that project’s lead builder has filed the same disclosure with Korea’s regulator seven times, on a rolling six-month clock, to say that its own construction contract is still under discussion. The eighth is due on 1 December 2026. Meanwhile a different Korean company signed ₩5.64 trillion of contracts on the same project — roughly $4bn — and filed them in a single afternoon. This is what a “nuclear winner” looks like when you open the order book.

🧭 Why This Article Exists

Because our own map of Korea’s nuclear export chain left the builders out, said so, and gave a reason we could go and test.

Earlier this month we published a map of Korea’s nuclear export chain organised by how many contractual steps each listed company sits from the entity that actually signs, with decodes of Doosan Enerbility (the fabricator), KEPCO E&C (the designer), KEPCO KPS (maintenance) and BHI (balance-of-plant equipment).

That map had a hole in it, and it said so. The words “Hyundai E&C” and “Daewoo” appear zero times in it. Its own honest caveat was that some participants are “invisible by construction” — their contracts fall below the disclosure floor, so “you will not find the contract.”

This article goes and looks. The answer turns out to be sharper than the caveat: one builder’s nuclear contract is perfectly visible, filed, dated and priced, with Korea Hydro & Nuclear Power named as the counterparty. The other builder’s contract is not invisible. It does not yet exist, and the company has said so in a filing.

🏢 What Do Hyundai E&C and Daewoo E&C Actually Sell?

Both sell construction labour and project management — overwhelmingly for apartments, offices and infrastructure inside Korea — and both describe nuclear power as one line item inside a diversified “Plant” division that is mostly oil, gas and petrochemicals.

Hyundai Engineering & Construction (KRX: 000720) was founded on 10 January 1950, seven months before the Korean War began, and listed on what is now the Korea Exchange on 22 December 1984. It is headquartered at 75 Yulgok-ro, Jongno-gu, Seoul. Its own annual report splits the business four ways: Civil (tunnels, bridges, roads, land reclamation, offshore wind), Building & Housing (apartments, public buildings, high-rises, data centres), Plant & New Energy (petrochemicals, gas processing, refining, large nuclear, SMR, solar, hydrogen electrolysis, ESS, transmission) and Other. It names its main customers as Korea Expressway Corporation, the Ministry of Land, the Public Procurement Service, LH, Aramco and KEPCO. (Source: DART FY2025 annual report, rcpNo 20260318001395.)

Daewoo Engineering & Construction (KRX: 047040) is the younger legal entity and the older brand. It was created on 27 December 2000 by spinning the construction division out of Daewoo Corporation, the trading house at the centre of the Daewoo Group collapse. It sits at 170 Euljiro, Jung-gu, Seoul. Its four segments carry the same names — Civil, Building, Plant, Other — and its Plant segment is defined in the filing as “Oil & Gas, LNG, Petrochemical, combined-cycle power, nuclear power generation and related businesses, renewables, industrial facilities.” (Source: DART FY2025 annual report, rcpNo 20260318001029.)

Read those two segment definitions carefully, because they are the single most important fact in this article. Neither company reports nuclear revenue or nuclear profit separately. Nuclear is bundled with refining and petrochemicals into a division, and the division is reported. Any figure you see for “Hyundai E&C’s nuclear business” is somebody’s estimate, not a disclosure.

📊 Where Does the Revenue Actually Come From?

From apartments. Housing and buildings are 54.3% of Hyundai E&C’s revenue and 68.4% of Daewoo E&C’s, while the Plant division that contains all nuclear work is 31.7% and 10.4% respectively — and most of that is petrochemicals.

Basis: consolidated, before elimination of ₩264.7bn of intra-group revenue. Shares are of the ₩31,327,660m segment subtotal; consolidated revenue after elimination is ₩31,062,912m ($22.43bn). (Source: DART rcpNo 20260318001395, “4. Sales and orders”, as at 31 Dec 2025.)

Basis: consolidated IFRS operating-segment note; shares are of consolidated revenue of ₩8,054,627m ($5.82bn); a −₩363,767m adjustment and −₩10,984m of unallocated operating loss are not shown as rows. Consolidated operating loss ₩815,431m (−$589M). (Source: DART rcpNo 20260318001029. Revenue and segment shares are from “II. Business — 2. Principal products and services”; operating profit by segment is from the consolidated operating-segment note, which is the only place the two are printed together.)

Hyundai E&C does not publish segment operating profit in the same form, so the two tables are deliberately shaped differently rather than forced into one. What the Daewoo table shows is worth sitting with: three of its four operating segments lost money in FY2025, and the only profitable one is property, leasing and hotels. The division that holds the nuclear business lost $91M.

🗓️ How Did Korea’s Two Most Famous Builders End Up Here?

One was absorbed into Korea’s largest carmaking group after a creditor bailout; the other was sold by a state bank to a regional housebuilder. Neither path was chosen by the companies.

Dates from each company’s DART FY2025 annual report unless noted. Ownership history is covered in the governance section below.

🌍 Are They Actually Leaders in Anything?

Yes — in Korean housebuilding. Daewoo E&C supplied more homes in Korea than anyone else for two years running, and Hyundai E&C leads urban redevelopment; neither claims, in any filing, to lead anything in nuclear.

Korea publishes an annual construction capability evaluation that ranks general contractors on a composite of performance, finances, technical capacity and credit. It is the industry’s official league table, and both companies report their placing in it.

All figures are as printed in each company’s FY2025 annual report (rcpNo 20260318001395 and 20260318001029). The “company-reported niches” row is the companies’ own characterisation, not an independently verified share, and we have not been able to test either claim. Neither report contains a market-share table for nuclear construction.

Take the ranking seriously in the direction it points. These are, by Korea’s own official measure, the third- and fourth-largest general contractors in the country, and Daewoo E&C has been the nation’s largest supplier of homes two years running. That is a genuine leadership position. It is a leadership position in building apartments.

Daewoo E&C’s nuclear claim is worth stating precisely because it is narrower than it sounds. The filing says it is the only Korean firm with experience across the whole nuclear life cycle — power reactors, research reactors, radioactive-waste disposal facilities and fuel-fabrication plants. That is a claim about breadth of past experience, not about current work or current share, and the order book bears that out: the experience is real, and today it amounts to one research reactor.

🔍 How Much Nuclear Is Actually in the Order Book?

Hyundai E&C has $1.04bn of named nuclear work against a $68.6bn consolidated contract balance — 1.52%. Daewoo E&C has $66M against $36.5bn — 0.18% — and it is an export research reactor, not a power plant.

Korean annual reports contain an order-book table that lists major projects by name, with the client, the base contract value, work completed to date and the balance remaining. It is the single most useful table in Korean corporate disclosure and it is almost never translated. Here is every line in either company’s table that names a reactor.

Basis: contract balance (계약잔액) at 31 Dec 2025, from each company’s order-book table. Hyundai E&C’s consolidated contract balance is ₩95,038,562m ($68.6bn) after eliminating ₩653,127m of intra-group work; Daewoo E&C’s order-book summary gives ₩50,596,818m ($36.5bn). Percentages in the text use those two totals. (Sources: rcpNo 20260318001395 and 20260318001029.)

Three things in that table deserve to be said out loud.

First, Barakah is finished. Korea’s only completed nuclear export — the four-reactor plant in the United Arab Emirates, contracted in March 2010 — shows $3.00bn of base value, of which 99.93% has already been recognised. What is left is $2.2M, and the scheduled completion date was 31 March 2026. Whatever Barakah does for Korea’s reputation, it does approximately nothing for Hyundai E&C’s future revenue. An investor buying the stock in 2026 for “the company that built Barakah” is buying a credential, not a cash flow.

Second, Shin-Hanul 3&4 is a nine-and-a-half-year job. The contract runs from 1 April 2024 to 31 October 2033. In the roughly twenty-one months to 31 December 2025, Hyundai E&C recognised ₩165,256m of it — 10.3% of the contract, and 0.53% of a single year’s revenue. A reactor build is not a revenue event. It is a thin, very long annuity.

Third — and this is the number that reframes the whole sector — the nuclear contract is small next to the apartments. The largest single line in Hyundai E&C’s domestic order book is the Banpo Jugong 1 redevelopment in Seoul at ₩3,931,885m ($2.84bn), which is 2.45 times the size of the entire Shin-Hanul 3&4 contract. And it is not alone. Reading the company’s own 2026 supply-contract filings:

Values as printed in each 단일판매·공급계약체결 filing on DART during 2026; the Shin-Hanul figure is the contract balance carried in the FY2025 order book, which is the conservative comparison (its signed value was higher — see below). All four are domestic property projects. Saudi petrochemical contracts filed in the same period are larger still but are excluded here because the filings do not separate the parent’s scope from Hyundai Engineering’s.

Hyundai E&C’s own annual report notes it passed ₩10tn ($7.2bn) of urban-redevelopment orders in a single year for the first time in December 2025. That is roughly six times the entire Shin-Hanul 3&4 contract, won in twelve months, in one domestic business line. If you want to model this company, model Korean apartment prices and Saudi Aramco’s capital budget. Nuclear is a rounding error with a very good press office.

🚧 Is Part of Hyundai E&C’s Business Currently Suspended?

An order to suspend it exists, and whether it is actually in force is not disclosed. A three-month suspension of Hyundai Engineering’s domestic civil and building construction licence — work Hyundai E&C’s own filing sizes at 17.8% of group revenue — was ordered for 5 August to 4 November 2026 after a serious industrial accident, and the company says it is seeking a stay.

This is the kind of item that is routine to file in Korean and essentially absent from English coverage. On 22 July 2026 Hyundai E&C filed a business suspension disclosure on behalf of its consolidated subsidiary:

Source: DART, 영업정지(종속회사의 주요경영사항), rcpNo 20260722800475, filed 22 Jul 2026. ⚠️ Whether the stay was granted is not disclosed in any filing we could find, so we do not know whether the suspension is actually in force. We are reporting the order and the company’s stated response, not the outcome.

It is not an isolated event. Hyundai E&C filed two serious-accident disclosures in 2026 — one on 26 June (amended 29 June) and one on 4 September. The September filing reports a worker killed on 3 September at the Shaheen ethylene facility PKG1 after the worker leaned out over an installed guard rail to pull a spray hose and lost his balance. It records one fatality and no injuries, and states that the Busan Regional Employment and Labor Office issued a partial work-stop order — “all fireproofing work and the like.” The filing names the site but gives no location and no contract value. (Source: DART rcpNo 20260904800020.)

Separately, and from a different document, the Shaheen project is one of the largest domestic lines in Hyundai E&C’s order book: the FY2025 annual report carries it at a base value of ₩2,438,912m with a completion date of 30 June 2026. (Source: DART rcpNo 20260318001395, order-book table.)

Korea’s Serious Accidents Punishment Act made this a balance-sheet risk rather than only a compliance one: it carries personal criminal liability for executives, and, through the Framework Act on the Construction Industry, suspensions of the licence a builder needs to bid. For a company whose bull case rests on winning large new contracts, an interruption to its ability to bid is a more immediate variable than any reactor.

🇨🇿 Did Daewoo E&C Actually Win the Czech Reactor Contract?

No. Daewoo E&C has filed seven separate disclosures over twenty-three months saying its own construction contract is not concluded; the most recent, dated 2 June 2026, says the contract is “currently under discussion” and sets the next update for 1 December 2026.

Korea has a disclosure type for exactly this situation: a clarification of a rumour or media report, which an issuer files to confirm, deny or qualify a press story, marked confirmed or unconfirmed, and which carries a binding re-disclosure deadline if the answer is “not yet.” Daewoo E&C has been filing the unconfirmed version about the Czech project since the story broke.

Source: DART, Daewoo E&C, 풍문 또는 보도에 대한 해명(미확정), rcpNo 20260602800524, filed 2 Jun 2026, and the list of related prior disclosures printed in it.

All seven respond to the same trigger: a 18 July 2024 Chosun Ilbo story headlined “Beat France in Europe — ₩24tn Czech nuclear order won” (about $17bn at today’s rate). Here is the operative part of the June 2026 filing, in our translation:

“To win overseas nuclear orders, ‘Team Korea’ was formed comprising Korea Hydro & Nuclear Power, the Company, and other nuclear public and private enterprises, and as reported, KHNP was selected as preferred bidder for the new Czech nuclear build.

Subsequently, while the construction contract was being concluded, Électricité de France filed an administrative objection in a Czech court, and the Brno regional court issued an injunction suspending conclusion of the contract until a ruling on the merits. On 4 June 2025 local time, the Czech Supreme Administrative Court overturned the injunction, and we have confirmed that KHNP and the ordering party concluded the main contract.

“At present, the conclusion of a construction contract between KHNP and the Company is under discussion. We will re-disclose at the point this is confirmed, or within six months.”

Source: DART rcpNo 20260602800524. The June 2025 version of the same filing used stronger language still — “nothing has been specifically confirmed.”

Separate the two contracts, because the English coverage does not. KHNP signed a contract with the Czech utility. Daewoo E&C has not signed a contract with KHNP. Team Korea is a bid consortium; being in it is a position, not a purchase order. The company itself draws exactly this distinction, and it draws it in the same year it tells shareholders the Czech project is a growth engine for the next fifty years.

What a signed contract on the same project looks like

The comparison that makes this concrete is not with a different industry. It is with the company sitting next to Daewoo E&C in Team Korea.

Both Doosan filings name 한국수력원자력(주) in the counterparty field and the Dukovany site in southern Czechia as the supply region, and both carry a contract date of 15 Dec 2025 running to 18 Apr 2038. (Sources: DART rcpNo 20251216801147 — ₩4,928,968,079,043 — and rcpNo 20251216801163 — ₩711,106,288,572.) A currency note, because this article promised not to re-convert a filing’s own figures without saying so: neither Doosan filing prints a dollar total. The contracts are denominated in euros and dollars (NSSS) and euros and Czech koruna (turbine generator), and each filing converts to won at its own contract-date rates — ₩1,728.72/€1, ₩1,472.50/$1 and ₩71.23/CZK1. The dollar figures here are therefore ours, converted from the filings’ won totals at this article’s ₩1,385/$1 so they sit on the same basis as every other dollar in this piece.

Read those three rows together. On one project, with one client, the equipment maker filed ₩5.64tn of signed contracts in a single afternoon in December 2025. Nine months later the construction lead has filed nothing, and its own latest disclosure says the contract is under discussion. That asymmetry is the whole investment case, and it is visible only in the Korean-language filings.

We tested the claim rather than taking it. We pulled and read every supply-contract filing either company made between 1 December 2025 and 20 September 2026 — 66 of them, amendments included — checking each for the Korean words for nuclear, nuclear plant, reactor, Czech, Dukovany, Shin-Hanul, SMR and research reactor. The result is zero. Not one of the 66 mentions any of them. What they do contain is apartments: redevelopment schemes at Sindaebang, Jeonnong, Sanggye, Cheonho, Cheongpa, Cheongju, Gunpo, plus two military-base contracts, a data centre, a road tunnel, and a research campus.

The only energy work Daewoo E&C disclosed in 2026 is gas-fired: main equipment for the Gongju natural-gas power station ($151M, from Korea Western Power, filed 28 Aug 2026) and for the Songsan Bitgreen cogeneration plant ($197M, from Korea South-East Power, filed 27 Jul 2026).

The precedent that tells you how this usually ends

Daewoo E&C’s own filing history contains a completed version of exactly this story, and it is the most useful thing in the document.

In October 2023 the Korea Exchange formally required the company to answer press reports that a “₩3tn ($2.2bn) Turkmenistan order” was imminent. Daewoo E&C answered seven times over eighteen months — 30 Oct 2023, 29 Nov 2023, 28 Feb 2024, 27 Aug 2024, 7 Oct 2024, 21 Oct 2024 and 18 Apr 2025 — and every answer was marked unconfirmed. The last of them states plainly that “the framework agreement dated 18 April 2025 does not constitute the main contract.” It also discloses something the original reports had missed: there were two fertiliser projects, and the company had lost the bid on one of them. (Source: DART, 조회공시 요구(풍문 또는 보도)에 대한 답변(미확정), rcpNo 20250418800618, which lists the six prior answers.)

The actual contract was disclosed on 26 May 2025, nineteen months after the first query: the Turkmenistan Mineral Fertilizers project. The filing prints the contract in dollars — USD 784,300,000, VAT included — and supplies its own won conversion, ₩1,081,079,120,000 at ₩1,378.40/$1, the mid-rate for 23 May 2025. That is 10.29% of revenue, and the contract date is 24 May 2025. (Source: DART rcpNo 20250526800282.) The press figure had been ₩3tn ($2.2bn). The signed contract was about a third of it.

A basis note, because the two figures differ and both are right: the FY2025 order-book table carries this same project at ₩978,478m, struck at the year-end accounting date rather than the signing date. (Source: DART rcpNo 20260318001029, order-book table.) The contract filing and the order book are not interchangeable, and this article uses the contract filing when it talks about what was signed.

We are not claiming the Czech project will follow the same ratio; that would be a guess dressed as analysis. We are pointing out that this company has a documented, disclosed, checkable history of large reported order numbers converting into much smaller contracts much later — and that the Czech number currently being quoted has not converted into anything at all.

📑 Where Do the Builders Belong on the Chain Map?

In a category that map deliberately excluded: Hyundai E&C signed directly with KHNP, which makes it a step-1 company — on a domestic reactor, and the map is a map of exports.

Our chain map sorts companies by what appears in the 계약상대 (counterparty) field of their contract filings. Step 1 means the filer signed directly with the ordering utility, and the map names three listed step-1 companies: Doosan Enerbility, KEPCO E&C and KEPCO KPS.

It also states its own boundary, before the table rather than after it: “The work is outside Korea. This is an article about an export chain, so domestic nuclear contracts with the same counterparty are excluded — and we should say plainly that the exclusion is not a size filter.” That is exactly right, and it is why the builders are absent from it. Hyundai E&C is not a fourth name the map missed. It is the first member of a category the map ruled out of scope — and on contract value it is larger than any of the three the map does list.

Applying the map’s own step-1 test to domestic work, Hyundai E&C’s trail is complete and public:

• 30 Nov 2023 — selected as provisional awardee for Shin-Hanul 3&4 main works. The filing gives the award at ₩3,119.6bn ($2.25bn) for the consortium, with the company’s own share “to be determined later.”
• 11 Dec 2023 — supply-contract filing. Value ₩1,715,754,987,100. Counterparty field: 한국수력원자력 주식회사 — Korea Hydro & Nuclear Power. Supply region: Uljin, North Gyeongsang. The filing states in its own notes that this amount “corresponds to the Company’s share (55%)” — so the 55% is disclosed, not derived, and it reconciles with the ₩3,119.6bn consortium figure announced eleven days earlier.
• 31 Dec 2025 — carried in the order book at ₩1,602,828m, ₩112.9bn (−6.6%) below the signed figure. The report warns that order-book values are struck on an accounting basis and may differ from the contract filings.

That is the decode a headline cannot give you: “Korea awards a ₩3.1tn reactor contract” means ₩1.72tn for this company, spread across nine and a half years.

It also demonstrates the disclosure mechanism the chain map described. Korea’s single-contract filing rule is proportional to the filer’s own recent revenue. The Shin-Hanul filing does the arithmetic for you: it prints a “recent revenue” of ₩21,239,081,840,150 and a ratio of 8.08%, noting that the revenue base is the FY2022 consolidated accounts — the last full year closed when the contract was signed. Comfortably over the floor, so it had to be filed, and it is visible.

The floor itself is lower than most readers will assume. For a company of this size the trigger is 2.5% of recent revenue, not 5%: across Daewoo E&C’s own 2026 filings the smallest disclosed contract prints a ratio of exactly 2.50% — ₩201,730,013,419 ($146M) for a Seoul road-tunnel job. And we do not have to infer the rule from the data, because a filer spells it out: Hyundai E&C’s 2022 NEOM disclosure states that “on the contract date the Company concluded a construction contract corresponding to 2.5% or more of the most recent fiscal year’s revenue“ — which is why it had to file at all. (Source: DART rcpNo 20220614800198.) Against FY2025 revenue of ₩8.05tn that puts Daewoo E&C’s floor near ₩201bn ($145M), half of what a 5% rule would imply. The zone a contract can hide in is smaller than the chain map’s caveat suggests, not larger. And in any case the company has not said its Czech contract is small. It has said it is under discussion.

Which raises the obvious question, and the answer is the cleanest structural fact in this article: Hyundai E&C is not part of the Czech project at all. The two builders are not competing for the same work — Daewoo E&C is in Team Korea for Czechia, and Hyundai E&C is not. Its Shin-Hanul contract is domestic, and its overseas nuclear activity runs on a different technology track — Westinghouse’s AP1000 rather than the Korean APR design — through engineering and study agreements rather than construction contracts. Its FY2025 annual report lists, among the year’s notable events, a design contract for Bulgaria’s Kozloduy 7&8 (November 2024), an Early Works Agreement on new Finnish nuclear (June 2025), and the Fermi America basic-design contract for four large reactors (October 2025). It also describes, in its strategy section, pursuing US reactor decommissioning and SMR work to accumulate experience. (Source: DART rcpNo 20260318001395.) Every one of those is design, study or partnership work. None of them is an EPC contract, and none of them appears as a line in the order book.

One more limit is worth printing, and it is narrower than we first assumed. Hyundai E&C filed its NEOM tunnel contract in Saudi Arabia on 12 June 2022 with the value temporarily withheld for commercial confidentiality — and then disclosed it when that reservation expired, in an amended filing on 29 July 2022 whose stated reason for amendment is simply “release of the commercial confidentiality undertaking”: ₩723,114,226,800, 4.00% of revenue. The contract is in Saudi riyals — the filing gives SAR 2,156 million at ₩335.32/SAR, and notes the amount is Hyundai E&C’s 35% share of a joint venture; the dollar equivalent of roughly $522M is ours, not the filing’s. (DART rcpNo 20220729800197.) So the signing was not a black box. On 12 March 2026 the client terminated the contract, citing “a request to terminate arising from the ordering party’s business restructuring.” That filing does withhold the termination amount, under a confidentiality clause in the contract — but it also states that settlement for work performed is complete and that “no financial loss has arisen to the Company to date.”

The sharper contrast sits in a field most readers never open. The 2022 signing filing withheld its value under a reservation with an expiry date printed on it — 31 July 2022 — and the value duly appeared two days early. The 2026 termination filing withholds its value under a reservation whose expiry field reads simply “–”. A deadline that was kept, and no deadline at all: that is the difference between a delay and a redaction. (Termination filing: DART rcpNo 20260313801644; original contract filing rcpNo 20220614800198.)

📈 Is the Business Getting Better or Worse?

Hyundai E&C is recovering from a heavy 2024 loss to a thin 2.1% operating margin; Daewoo E&C went the other way, from a 3.8% margin in 2024 to a −10.1% operating loss in 2025, and its equity fell by a fifth.

Basis: consolidated (CFS) throughout, from each company’s DART FY2025 filing. “Net profit (total)” is profit for the period including non-controlling interests, so the two companies are on the same basis. The split matters very differently for each: at Daewoo E&C the FY2025 loss is ₩916,080m in total, of which only ₩3,816m belongs to non-controlling interests, so the total and the attributable readings are practically the same number. At Hyundai E&C the FY2025 profit is ₩559,068m total and only ₩373,147m attributable — a gap of 33.3%. See the caution immediately below. (Sources: rcpNo 20260318001395, 20260318001029; DART fnlttSinglAcntAll FY2025 CFS.)

💰 What Do You Actually Pay for These Two Today?

Hyundai E&C trades at 37.2× attributable earnings and 1.37× book; Daewoo E&C has no P/E because it lost money, and at 2.24× book it is the most expensive listed builder in Korea while being the only unprofitable one.

Prices are KRX settled closes for 18 September 2026; market caps are KRX’s own, same date. Revenue, operating margin, equity and liabilities are FY2025 consolidated from each company’s DART annual report. P/B = market cap ÷ total equity including non-controlling interests. Samsung C&T is included for scale but is a trading, fashion and holding conglomerate whose equity includes large listed stakes, so its multiple is not like-for-like. (Sources: KRX Open API; DART fnlttSinglAcnt FY2025 CFS.)

Sort that table by P/B and the ranking is the exact inverse of the ranking by profitability. The cheapest builder on book value earns a 6.0% margin; the most expensive one lost 10.1% of revenue at the operating line and carries the highest leverage in the group. Daewoo E&C trades at 3.9× GS E&C’s price-to-book and 4.7× HDC’s.

There is a legitimate bull answer to this: book value is backward-looking, FY2025 was a kitchen-sink year, and a builder emerging from a loss cycle should be expensive on trailing numbers. That argument is real. It is also an argument about Korean housing and cost inflation — not about reactors.

And the year-to-date number is hiding the most important price fact

KRX settled closes. Share counts were unchanged between 30 Dec 2025 and 3 Mar 2026 at 415,622,638 for Daewoo E&C and 111,355,765 for Hyundai E&C, so neither return is a stock-split artifact; Daewoo’s count fell to 410,907,638 only in March, through the cancellation described below. (Source: KRX Open API daily trade data.)

Both stocks peaked in April, lost 56.7% and 69.9% respectively into a trough on the same day, 30 July 2026, and have partially recovered. Daewoo E&C is up 395% for the year and down 49.1% from its own high at the same moment. Both statements are true; quoting only the first is the more common error.

Note also what that path is not. A 9.6-year EPC contract does not gain and then lose two-thirds of its value in four months. Nothing happened to either company’s contracted cash flows in that window. These prices are moving on expectation, and expectations reverse.

🏷️ Which Ticker Are You Actually Buying?

Hyundai E&C has two listed lines — common stock 000720 and a preferred share 000725 that trades at 42.4% of the common price — while Daewoo E&C has only one, and the group company most exposed to Hyundai E&C’s order book, Hyundai Engineering, is not listed at all.

Sources: KRX Open API, 18 Sep 2026; DART rcpNo 20260318001395 (subsidiary schedule, dividend table). Neither company has a US ADR; access routes are covered in our guide to buying Korean stocks as a foreign investor and our list of Korean ADRs.

🌏 Is Foreign Money Buying or Selling?

It depends entirely on the window, and for Daewoo E&C the sign actually flips: foreigners were net sellers over the last twenty sessions and net buyers over the last sixty-five.

Method: each session’s net share count valued at that session’s own KRX settled close, summed, then converted once at ₩1,385/$1. This is an approximation from settled net positions, not execution prices; it differs slightly from our weekly Foreign Flow Watch series, which converts at each session’s own FX rate. The 65-session window is the full span of our flow dataset for these names (18 Jun – 18 Sep 2026). (Source: our own investor-flow dataset, built from KIS Open API; prices cross-checked against KRX Open API.)

Look at the two Daewoo E&C rows that disagree. Over twenty sessions foreigners sold $60.6M and the stock rose 18.71%. Over sixty-five sessions they bought $13.2M and the stock fell 17.60%. A writer with one window and a deadline can produce either “foreigners are abandoning Korean builders” or “foreigners are quietly accumulating.” Both would be sourced. Both would be worthless. Where the sign flips, the flip is the story. How these flows are constructed, and why they mislead as often as they inform, is set out in our pillar on how foreign and institutional flows move Korean stocks.

The other pattern worth naming: over sixty-five sessions domestic institutions bought $250.2M of Hyundai E&C while retail sold $193.5M, and the stock still fell 6.31%. Institutions were the marginal buyer of both names over the longer window.

📉 Why Did Both Stocks Jump More Than 22% in a Week While Foreigners Sold?

The whole Korean nuclear complex moved together in the same sessions while foreigners sold all of it, and the one nuclear-related filing in that window is Hyundai E&C telling the regulator on 24 August that it had finished a design study and hoped to sign an EPC contract.

In our seventh Foreign Flow Watch issue we printed a table of names where flow and price disagreed over the five sessions of 24–28 August 2026. At the time we treated it as a list of counterexamples. Reading it back with this article’s lens, the list has a shape we did not name:

Foreign net covers the five sessions 24–28 Aug 2026. The price column uses this series’ standard weekly window — the prior week’s last close to the current week’s last close — so it spans one session more than the flow column. That is deliberate and is how the series has measured weekly moves since issue #7. (Source: Foreign Flow Watch issue #7, published 29 Aug 2026.) Doosan Enerbility added the largest market value on the KOSPI that week on foreign net selling of $4.6M — a flow too small to register on any ranking.

Four of the eight names in that week’s disagreement table are nuclear-chain companies, and they are the top four by gain. Our own daily data shows the same thing at higher resolution: Hyundai E&C and Daewoo E&C posted their two largest single-day gains of the period on the same two dates — +15.79% and +14.11% on 31 July, then +14.87% and +11.80% on 25 August — and troughed on the same day, 30 July.

Two stocks with different segment mixes, different balance sheets, different controlling shareholders and no shared contracts do not move 12–16% together on the same days for company-specific reasons. Something re-priced the whole complex, and it was not foreign buying, because foreigners were net sellers of every name in the table.

The one filing in that window — and what it actually says

Both companies filed routine contract disclosures that week — Daewoo E&C alone filed four between 19 and 28 August, all of them domestic property and power work. But there is exactly one nuclear-related disclosure from either name in the days around 25 August, and it is worth reading in full because of how modest it is. On 24 August 2026, Hyundai E&C filed a clarification — again of the “unconfirmed” type — responding to a Seoul Economic Daily story from 27 October 2025 headlined “Hyundai E&C joins first US large nuclear build — a ₩60tn jackpot?” (about $43bn) The company’s answer, in our translation:

“The Company has completed the basic design services for US large-reactor construction contracted with FERMI America on 24 October 2025, and is reviewing specifics including feasibility, safety and schedule with the goal of concluding an EPC contract. We will re-disclose when specifics are determined, or within six months.”

Source: DART, Hyundai E&C, 풍문 또는 보도에 대한 해명(미확정), rcpNo 20260824800602, filed 24 Aug 2026. Next scheduled re-disclosure: 23 February 2027. It is the fourth filing in that chain, after 27 Oct 2025, 26 Nov 2025 and 25 Feb 2026.

Strip the vocabulary and the structure is identical to Daewoo E&C’s: a press number in the tens of trillions of won, a company confirming only that preparatory work is finished, an explicit statement that the real contract is a goal rather than a fact, and a six-month clock. The following session, the stock rose 14.87%.

We are not claiming the filing caused the move, and the data argues against a simple story. On 25 August itself, KEPCO E&C rose 20.27% and BHI 13.80% — both more than Hyundai E&C’s 14.87% — and neither has any Fermi America relationship; Doosan Enerbility rose 10.55% and Daewoo E&C 11.80% (KRX settled closes, 24→25 Aug 2026). A complex-wide re-rating that happens to coincide with one company’s filing is not the same as that filing moving four stocks. What the filing does establish is the quality of the news available at that moment: a completed design study, not an order.

🏛️ Who Controls These Companies, and Who Gets the Profit?

Hyundai E&C is controlled by a group of Hyundai Motor affiliates holding 34.92% between them, with the National Pension Service the largest single outside holder at 11.08%; Daewoo E&C is controlled outright by the unlisted Jungheung group, whose two companies hold 50.75%.

The three Hyundai Motor Group affiliates hold 38,879,000 shares, 34.92% between them. Rows above are individual holders, not a sum. (Source: DART rcpNo 20260318001395, “Holders of 5% or more” and “Minority shareholder status”, as at 31 Dec 2025.)

The two Jungheung companies hold 210,931,209 shares between them. The filing’s own total of 210,946,209 — 50.75%, an outright majority either way — also includes 15,000 shares held by the then chief executive. Rows above are individual holders, not a sum. Common stock, as at 31 Dec 2025. (Source: DART rcpNo 20260318001029.)

The structures produce different risks. At Hyundai E&C, three chaebol affiliates hold a third of the company, and the filings also record it buying steel from Hyundai Steel, rolling stock components from Hyundai Rotem and logistics from Hyundai Glovis — all named as group affiliates — while building the group’s own Global Business Center headquarters under a ₩1.79tn ($1.29bn) contract with Hyundai Motor, Kia and Mobis as joint clients. Related-party revenue is not hidden; it is simply large. At Daewoo E&C, a single unlisted family group holds an outright majority, which removes any question of who decides and equally removes any contestability.

What each company actually returned to shareholders in 2026

Hyundai E&C paid ₩800 ($0.58) per common share, a 24.10% payout ratio and a yield the report puts at 0.8%; Daewoo E&C paid nothing, and its two 2026 “share cancellations” together retired 1.14% of the company — the second of them was 0.005%.

Hyundai E&C filed a Value-Up plan on 28 March 2025 and an implementation update on 17 December 2025. Its stated targets, which the filing explicitly labels as forward-looking information that may change:

• Revenue of ₩40tn ($28.9bn) or more by 2030 (FY2025: ₩31.06tn / $22.43bn)
• Operating margin of 8% or more by 2030 (FY2025: 2.1%)
• ROE of 12% or more by 2030; total shareholder return of 25% or more over 2025–27
• A minimum dividend of ₩800 ($0.58) per common share — which is exactly what it paid for FY2025
• Strategy described as “growth centred on the energy industry”

Source: DART, 기업가치 제고 계획(자율공시), rcpNo 20251217800377, filed 17 Dec 2025. These are company targets and forecasts, not commitments.

Read that as management has written it. The company’s own plan says the 8% margin arrives in 2030. It is running at 2.1% now. The energy pivot is being sold as a five-year transformation by the company selling it, while the stock already carries 37.2× trailing attributable earnings. Whatever else it is, it is not a cheap way to wait.

Daewoo E&C’s side is where the primary documents earn their keep. It paid no dividend for FY2025 — unsurprising, given a ₩916bn attributable net loss. It did, however, file two treasury-share cancellations, and if you read the headlines rather than the filings you would take them for the same kind of event.

Both cancellations retire shares without reducing paid-in capital. (Sources: DART rcpNo 20260304801486 and rcpNo 20260910800468.) Cross-check: each filing’s stated value divided by its share count gives ₩8,900 and ₩19,450 — the KRX settled closes on 3 Mar and 9 Sep 2026, the day before each board meeting. The arithmetic reconciles exactly.

The September filing announces the retirement of 21,918 shares worth about $310,000 — five thousandths of one per cent of the company — and the shares in question are odd-lot fragments created by a capital reduction resolved nineteen years ago, which had been sitting in treasury ever since. It is a housekeeping entry. In an English headline it reads identically to a buyback.

There is a real point underneath the curiosity. The two cancellations together retire 4,736,918 shares, and the FY2025 annual report puts the company’s entire treasury holding at exactly 4,736,918 shares. After 30 September 2026, Daewoo E&C’s treasury stock is zero. No further cancellation is possible without first buying shares back — from a company that posted a $589M operating loss and whose liabilities are 284% of equity. For what a cancellation is worth when it is real, see our decode of Celltrion retiring 4.2% of its market capitalisation, and our guide to Korea’s Value-Up programme.

📚 Lingo Check

Definitions drawn from our editorial glossary; new terms added back to it.

🎯 Why It Matters for K-Export Stars

This is the clearest case we have published of a number that exists in English and a fact that exists only in Korean, pointing in opposite directions.

This site exists because there is a gap between what Korean companies file in Korean and what reaches an English-speaking investor. Usually that gap is a detail: a footnote, a share-count bridge, a counterparty field. Occasionally it is the whole story.

“Korea wins ₩24tn Czech reactor order” has been in circulation in English since July 2024. The construction lead has since told Korea’s regulator seven times, in Korean, that its own contract is not concluded — most recently on 2 June 2026, with the next update scheduled for 1 December 2026. Its 2026 contract filings, 66 of them since December 2025, contain no nuclear work whatsoever. Nobody hid any of this. It sits in documents anyone can download, in a language most of the buyers of these shares do not read.

The same is true of the three-month licence suspension ordered against a Hyundai E&C subsidiary, covering domestic work the company itself sizes at 17.8% of group revenue. Whether the order is in force today is something no filing tells us — but its existence, its size and its dates are all in a document filed in July. That is a larger, nearer and more quantified fact than anything in either company’s nuclear story.

That is the entire thesis of this publication, and it is why we built the chain map from contract filings rather than from press releases. The map had a hole where the builders should have been. Going to look filled it in — and what was in the hole was more interesting than what we expected to find there.

Disclaimer: This article is for informational and educational purposes only and is not investment advice, a recommendation, or an offer to buy or sell any security. Figures are drawn from primary filings and exchange data as dated in the text and may be superseded. Prices, valuations and flow data are point-in-time and go stale quickly. Do your own research and consider your own circumstances before investing. See our corrections policy.