1942Okubo works — later Kobelco Construction Machinery
Kobe Steel began in September 1905 as an acquisition. Suzuki Shoten, a Kobe trading house dealing in sugar and camphor that would grow into Japan’s largest trading company, bought a small foundry at Wakinohama run by Kobayashi Seiichiro and renamed it Kobe Steel Works. The buyer was Kaneko Naokichi, the head clerk who ran the firm after the founder’s death. In June 1911 the works was separated out as a joint-stock company with capital of ¥1.4 million, and a retired navy naval-constructor rear admiral, Kurokawa Yukuma, was brought in as its first president.
The name said steel, but the business was never one material. Because the parent was a trading house, the point was to make domestically what Japan was importing: within a year of independence, in March 1912, the company produced Japan’s first drills, breaking into cutting tools that had until then come entirely from abroad. A 1,200-tonne press installed in 1913 gave it large forgings such as ship crankshafts; a torpedo air compressor perfected in 1915 won it essentially the navy’s whole requirement. Machinery followed in 1915, steel rolling in 1916, and copper and brass at the Moji works in 1917. In 1921 it bought shipyards from Teikoku Kisen and added shipbuilding and electrical equipment.
Then the ground gave way. In April 1927 Suzuki Shoten failed in the financial panic, and Kobe Steel survived as an independent company with no zaibatsu group behind it. Outside the protected orbit of Mitsui, Mitsubishi, Sumitomo and the state-backed Nippon Steel, it could only add businesses of its own: high-grade special-steel wire rod from 1933, which grew to over 30% of national output; aluminium in 1937; Japan’s first welding rods in 1940; new works at Chofu (1939) and Okubo (1942), the latter the ancestor of today’s Kobelco Construction Machinery. Wartime demand swelled it to some 22 sites and 70,000 employees; almost every plant was bombed. It cut sixteen works to twelve, and in November 1945 was the first in the industry to tap iron again.
1958Japan’s first plant export — a fertiliser plant for East Pakistan
1959Nadahama No.1 blast furnace lit — an integrated producer
1965Absorbs Amagasaki Steel and its two blast furnaces
1970Kakogawa Works begins integrated production
1973Kakogawa No.2 furnace — six million tonnes a year
The company relisted and rebuilt. Shares were listed in Tokyo, Osaka and Nagoya in May 1949; in August it spun two subsidiaries out of its non-ferrous and electrical plants, then reversed course in 1957 and folded the alloy business back in. Around that core it kept widening — the Takasago works for machinery in 1953, a joint venture with America’s Pfaudler in 1954, a stake in Nippon Koshuha Steel in 1955 — but in steel itself it had a structural weakness. With the arrival of the basic-oxygen converter, integrated mills that took ore all the way to finished steel became decisively cheaper, and Kobe Steel had no blast furnace of its own. It bought its pig iron from Amagasaki Steel, and alone among the six major producers it had no hot strip mill. As a wire-rod and forging house it simply could not reach the growing market for sheet.
Fixing that took fifteen years and three moves. From 1957 it reclaimed land at Nadahama and spent ¥37 billion on two blast furnaces; the first was lit in January 1959, making Kobe Steel an integrated producer fifty-four years after its founding. Rod and bar alone were still not a full-line mill, so in April 1965 it absorbed Amagasaki Steel and its two furnaces outright. Finally it built a new integrated works on reclaimed land at Kakogawa, which began integrated production in March 1970 and, with its second furnace in 1973, reached six million tonnes a year — plate, hot-rolled and cold-rolled sheet at last alongside the rods.
The rest of the group grew in parallel: Moka for aluminium in 1969, Saijo in 1970, Fukuchiyama in 1975. Machinery went abroad early, winning a fertiliser plant for East Pakistan in 1958 in what was Japan’s first plant export. When the oil crises thinned steel demand in the 1970s, engineering was where the company pushed — by fiscal 1980 it was over 10% of group sales — and in building Qatar Steel’s mill in 1978 the engineering division adopted an American process for reducing iron ore with natural gas. It thought the technology superior. That judgement led straight to the 1983 acquisition.
1976Composite management, and the refusal to merge
Revenue (¥ bn, bars) · net margin (%, line)
Source: securities reports
FY1976 · unconsolidated
Revenue$2.5B
Net income$11M
Net margin0.4%
→
FY2001 · consolidated
Revenue$11.3B
Net income$53M
Net margin0.5%
1983Acquires Midrex; buys into Yutani Heavy Industries
1990Semiconductor joint venture with Texas Instruments
1995Great Hanshin earthquake — ¥109.6bn of damage
1998DRAM venture handed to Micron Technology
1999Mizukoshi Koshi’s “Ten Project”; independence over merger
July 1983 brought two purchases that shaped the next forty years. Kobe Steel took over the loss-making American firm Midrex, acquiring direct-reduced-iron technology it would never once use in its own mills — the process makes feedstock for electric furnaces, and Japan ran on blast furnaces — and it bought into Yutani Heavy Industries, its entry into construction machinery, formalised as Kobelco in 1986. After the 1985 Plaza Accord drove the yen up and Asian rivals closed in, the company cut unprofitable lines from late 1986, then swung to expansion in 1989: Kobe Steel USA in 1988, the USS/KOBE bar and rod joint venture in 1989, alliances with Alcoa in aluminium and Texas Instruments in semiconductors in 1990.
The 1990s punished it. The Great Hanshin earthquake of 17 January 1995 wrecked wire-rod and steelmaking shops and the head office, killed three employees and cost ¥109.6 billion; accumulated losses reached ¥85.7 billion and the dividend did not return until fiscal 1997. The semiconductor venture, a commodity DRAM fab in Hyogo built with Texas Instruments, was hit by the price collapse from fiscal 1996. Rather than quit, Kobe Steel swapped partners to Micron Technology in June 1998 — a buyer that would use the existing plant, cutting the required investment from about ¥47 billion to ¥20 billion — yet the electronics division still lost ¥9.3 billion at the operating line in fiscal 1998, the group lost ¥23.3 billion, and the dividend was suspended again.
That is the setting for the decision the company is best known for. Mizukoshi Koshi became president in April 1999, introduced a company system, and put every peripheral, synergy-free business up for sale under a “Ten Project” to be finished within the year; construction machinery was consolidated into Kobelco, and the US bar and rod venture was merged away until Kobe Steel’s stake fell to 15% and left the consolidation. Meanwhile the industry consensus held that five integrated steelmakers were two too many. Mizukoshi dismissed merger as a fashion, argued that Kakogawa and Kobe run as one gave him six million tonnes and no real scale disadvantage, and stayed independent. He had the numbers: the highest consolidated operating margin in steel among the five, a quarter of the domestic wire-rod market, and none of the three loss-making product lines that burdened his rivals. The price of that refusal was that the steel cycle would hit the P&L undamped.
2013Decision to shut the Kobe blast furnace; equity raise
2017Quality-data falsification disclosed; shares fall 43%
2019Midrex and ArcelorMittal agree on hydrogen direct reduction
2022Midrex wins the first commercial 100%-hydrogen plant
Having refused to merge, Kobe Steel had to build its own shock absorber against the steel cycle, and it chose electricity. In March 1996 it bid into Kansai Electric’s wholesale power auction with a coal-fired station inside the Kobe works — about ¥200 billion, of which ¥165 billion was raised as project finance — and units came on stream in April 2002 and April 2004 for roughly 1.4 million kilowatts. The 2000s were kind: Chinese demand carried the group to a peak of ¥2,132.4 billion in sales and ¥202.4 billion in operating profit in the year to March 2008, before the financial crisis pushed it to a ¥31.4 billion net loss a year later. Steel then lost roughly ¥120 billion cumulatively over the five years from fiscal 2009, producing 7.61 million tonnes in fiscal 2013 — a sixth of Nippon Steel & Sumitomo Metal.
Kawasaki Hiroya took over in April 2013 with twelve loss-making years out of the previous twenty-five behind him, and made the centrepiece of his plan the shutdown of the Kobe works’ blast furnace — the very furnace lit in 1959 — consolidating ironmaking at Kakogawa for ¥75 billion a year of savings, financed in part by a $851.4M (¥83bn) equity offering. Then, on 8 October 2017, the company disclosed that it had shipped aluminium and copper products as conforming when they did not meet contract specifications. The falsified volume was about 4% of the divisions’ annual shipments, some ¥13 billion; the share price fell 43% in four days; over 600 customers were affected. Kawasaki attributed it to management that watched divisional profit and nothing else. The president and vice-president resigned, dozens of managers were disciplined, Yamaguchi Mitsugu became president in April 2018, and a summary court fined the company ¥100 million in March 2019.
What rescued the story was the technology bought in 1983 and never used. Midrex holds a standing share of about 60% of direct-reduced iron output worldwide; combined with hydrogen its process can cut ironmaking CO₂ by 20–40%, and buyers arrived once decarbonisation became policy. Kobe Steel signed a joint development agreement with ArcelorMittal on hydrogen-based direct reduction in September 2019, and in October 2022 Midrex won the world’s first commercial 100%-hydrogen plant from Sweden’s H2 Green Steel. Power reached six units — four at Kobe, two at Moka — by February 2023. Earnings recovered to ¥2,543.1 billion in sales and ¥109.6 billion in net profit in the year to March 2024, when Katsukawa Yoshihiko succeeded Yamaguchi. The businesses long dismissed as an unfocused collection had become the reason the company still had options.
In a 1964 lecture, managing director Minato Shizuo answered that “setting the future aside, up to now we have had no need of it, so we feel no delay” — and in the same session said the time was about right to merge with Amagasaki Steel. He would not concede that lacking a hot strip mill was a weakness, yet he was already arranging to bring his pig-iron supplier in-house. For a company that held 40% of the country’s special wire rod and earned its living from welding consumables and machinery, having no blast furnace was a weakness and, at the same time, an agility — capital that did not have to sit idle.
Kakogawa was the purchase that spent that agility. Enormous sums went in from the 1968 reclamation through the second furnace in 1973, with a further several hundred billion yen budgeted for the third phase, and completion coincided with the oil shock — executive director Inoue Kozaburo spoke of the third phase amid a crisis without precedent in the postwar era. Even so, owning a mill in Harima that could make sheet is what left room, in the 2010s, to fold up the furnace at Kobe. Whether an investment was right can only be judged not by what happens just after start-up, but by what it leaves behind.
Kobe Steel has never once used the process it bought in 1983 in its own mills. Midrex’s method of reducing iron ore with natural gas makes feedstock for electric furnaces; there was no demand for it in a country running on blast furnaces, and its siting was limited to places with natural gas. That the company could buy it anyway was because the machinery and engineering division, which had built Qatar Steel’s mill itself, looked at the process as a seller rather than a user. It was a different arena, chosen by a company that could not match Nippon Steel or NKK on blast-furnace scale.
The forty years in between were not smooth. ITmk3 saw its first commercial plant start up in 2010, yet through 2011 it ran intermittently because of coal-conveyor trouble and never reached its rated 500,000 tonnes a year. Per-unit scale fell short of a blast furnace, and president Sato Hiroshi himself listed “whether a market exists” as an open question. Midrex was only regarded as the front-runner once decarbonisation arrived; to say the 1983 acquisition foresaw this would be hindsight. That the company held on to a technology with no visible use for forty years is itself what this decision achieved.
In the same year it let go of a semiconductor business that had earned ¥20 billion of operating profit the period before, Kobe Steel committed ¥200 billion to coal-fired power. The character of the decision shows in that simultaneity of exit and vast investment. Having decided not to join a merger, it had no choice but to build its own earnings source to absorb the swings of the steel market. A regulatory window that procured generating capacity by auction, operating experience with in-house power going back to 1959, and an urban site equipped with a coal berth — the three came together only because this was a company that had persisted with composite management.
Yet creating stable earnings also invited instability of another kind. Concentrating 2.7 million kilowatts into four units meant that one unit going down could move the whole group’s guidance. For the year to March 2026, an extended periodic inspection on unit 3 was cited as a reason power earnings would fall. A business chosen out of distaste for the steel market’s volatility has produced a structure in which results ride on the availability of a single machine. The swing itself was not removed; the kind of swing was exchanged.
Borrowed technology goes stale unless you keep borrowing
As a way of touching leading-edge manufacturing technology without owning a single patent of its own, being folded wholesale into TI’s production system was, at the time, a rational choice. An 80% yield was the proof of it, and president Kametaka Sokichi had grounds for saying he had no intention of running a losing business. But borrowed technology goes stale unless you keep borrowing. From 0.28 microns to 0.18, and then to 300-millimetre wafers, a staircase of investment invisible at entry piled up over ten years.
The exit was decided at a moment when the venture had climbed to ¥5.1 billion of recurring profit, with accumulated losses in sight of being erased within two years. That it was decided not on near-term profitability but on whether the next step up could be climbed is what marks this judgement. President Mizukoshi Koshi made “synergy” the yardstick because he could find no reason to keep pouring steel’s earnings into a business to which neither steel’s technology nor its people transferred. What survived that same yardstick was power generation, which could use the operating experience of a steelworks directly. The technology borrowed in semiconductors left with the shares; the technology used in power stayed. What separated success from failure in diversification was the difference between technology whose value vanishes when you stop investing and technology that remains in your hands when you do.
The shares were held in order not to merge. At the comprehensive alliance announced in December 2001, vice-president Mitsutake Norifusa stated flatly that there would be no capital relationship and no change to an independent course; about a year later Kobe Steel exchanged capital with Nippon Steel and Sumitomo Metal anyway. A few percent was far too small to reach into a partner’s management, and probably just enough to secure someone to consult if a takeover proposal arrived. For a company with a ¥1.3 trillion market capitalisation, it was likely the smallest sum that would buy continued independence.
That umbrella, however, never once opened. Mittal Steel never reached for Kobe Steel, the three-way memorandum’s joint studies never moved, and half the stake was sold in 2014 and the remainder in 2025. In both cases the reason was the balance sheet and capital efficiency, not the content of the alliance. President Inubushi Yasuo had said that competitiveness in steel rests on the mills; twenty-three years on, what remained was not the shares either, but the practical business of trading iron units back and forth.
Four and a half years of building in order to close
The decision to stop a blast furnace could not be executed on the day it was made. Four and a half years passed between the resolution of May 2013 and the actual banking of the fire because other things had to come first: some ¥65.5 billion of work at Kakogawa — a sixth continuous casting shop, twice the blooming-mill reheating capacity, added degassing and refining furnaces — and the time needed to re-obtain quality approvals from the automakers. And the company raised part of the money from the market, in a $851.4M (¥83bn) public offering in fiscal 2013. The decision to fold up appeared first as an investment.
Yet the ¥15 billion a year the consolidation delivered was not enough to bring a production cost of a little over ¥80,000 a tonne down to JFE Steel’s ¥65,000. The scale disadvantage that president Kawasaki Hiroya described as leaving the company liable to “be wiped out” survived the move to a single site. By 2024 even the two furnaces kept at Kakogawa were under study for reduction to one blast furnace and one electric furnace. The line of furnaces lit at Nadahama in 1959 did not come to rest with the 2017 consolidation; it remains in the middle of a contraction.
A culture of profit first, under a distinguished name
At the root of this misconduct lies the way a composite manufacturer’s separated divisions had each been run against the single yardstick of their own profit. Quality assurance and quality audit sat inside divisions chasing that profit, so that delivery and margin pressure reached the inspection floor — an arrangement only articulated as a structural problem once it was pointed out from outside. That chairman and president Kawasaki acknowledged he had “only been looking at whether profits were there” was candid, and at the same time a measure of the gap between a company praised as first-rate and what was being left unattended beneath it.
Resignations and fines are a necessary settlement, but they do not by themselves replace a corporate culture. In the past too, each scandal brought an apology from the top and a pledge of prevention, and still the disposition that put divisional profit first remained. Whether quality assurance can be separated from the divisions, whether people on the floor can voice dissent — whether the distinguished name survives depends on the plain, long rebuilding that follows the single decision to disclose, and on how well it is done.
Each heading links to the full Japanese analysis — background, decision and outcome, with sources.
This is a condensed English edition. The full, source-by-source history — with the detailed narrative, financial tables, shareholders and executives — is maintained in Japanese: 日本語版（詳細）— Kobe Steel full history in Japanese →
Kobe Steel, Ltd. — 有価証券報告書 (annual securities reports), including the corporate-history and officers sections.
A Comprehensive History of Japanese Companies — 『日本会社史総覧』 (Toyo Keizai, 1995).
Eighty Years of Companies and Banks, vol. 2 — 『会社銀行八十年史 第2篇』 (Toyo Keizai, 1955), the Kobe Steel and steel-industry chapters.
A History of Enterprise: One Hundred Years of Meiji — 『企業の歴史：明治百年』 (Keizai Shunjusha, 1968), “Kobe Steel Works”.
Securities Analysts Journal — 証券アナリストジャーナル, vol.2 no.7 (1964), Minato Shizuo, managing director, on Kobe Steel; vol.12 no.12 (1974), Inoue Kozaburo, executive director, on composite management in steel, heavy machinery and non-ferrous metals.
Shukan Toyo Keizai — 週刊東洋経済: 18 Jan 1997 (new direct-reduction processes); 6 Nov 1999 (Mizukoshi Koshi interview); 11 Mar 2000; 26 Aug 2000 (the Micron deal); 22 Sep 2001 (the IPP investment); 4 Jun 2005 (the new iron-source division); 24 Oct 2009; 9 May 2014 (the blast-furnace shutdown); 2 Dec 2017 (the falsification, governance and prior scandals); 19 Nov 2022 (decarbonised ironmaking).
A History of Japanese Industry, vol. 3: Energy and Resources — 『日本産業史3 エネルギー・資源』 (Nikkei Inc., 1994).
Nikkei — 日本経済新聞: 6 Mar 2018 (resignations over the quality fraud); 13 Mar 2019 (the ¥100m fine); 18 Jul 2019 (close of the US Department of Justice inquiry).
Kobe Steel, Ltd. — corporate releases: wholesale power entry (19 Mar 1996); 50 years of integrated production (Dec 2009); Midrex–ArcelorMittal hydrogen direct-reduction agreement (17 Sep 2019); Kobe Power Plant overview; FY2024 results presentation (12 May 2025).
This page is provided for general information only and is not investment advice, nor a recommendation to buy or sell any security.
Figures are compiled independently and include our own estimates, approximations and machine-processed data; we make no warranty as to their accuracy or completeness.
Sources are primarily each company’s securities reports and other public filings, but errors and omissions may remain.
Any use of this information is at the reader’s own risk. Past performance does not indicate future results.
Company names, logos and other marks belong to their respective owners.
Data API
Kobe Steel’s history, financials, executives and
shareholders are published as static JSON — no key, plain GET.