Company history

Japan’s semiconductor distribution in 1972 ran through the sales channels of the big electrical makers, and an independent had little room to push in. Kamiyama Haruki, twenty-six, founded Japan Macnics that year in an office above a bakery, and the only opening left to him was the one the large trading houses ignored: small American chipmakers with revenues in the tens of millions of dollars and no partner in Japan.

Selling those parts on a datasheet alone offered no differentiation, so Macnica sold engineering instead — placing its own engineers inside the customer’s design team and working the circuit out with them. That human service was the margin. The company moved to Kawasaki in 1975 to sit close to the electronics makers of the southern Tokyo belt, developed its own JET101 Ethernet transceivers in 1983 as the standard was being written, and in the late 1980s took on the Japanese market for two Silicon Valley startups then barely known — Xilinx and Altera.

That last bet defined the next thirty years. When FPGAs grew from a curiosity into a category, the volume flowed straight through the distributor who had carried them when they were nothing. In 1990 Macnica built a head office in Yokohama, where it has been based ever since.

Listed in 1991, Macnica spent the next two decades on a single model: near-exclusive relationships with emerging suppliers on one side, resident design support on the other. Being the supplier’s only route into Japan let it share some of the inventory risk that has bankrupted independent distributors; being inside the customer’s design work let it forecast demand more accurately than a catalogue seller could. The result was an integrated-circuit business that cleared ¥200bn a year through the 2000s.

It was also a business tied to one cycle. Macnica’s customers — industrial equipment, telecoms gear, factory automation — moved their chip buying with their capital spending, and in a downturn falling volume and inventory write-downs arrive together. Kamiyama treated that fragility as unfinished business, and the answer he built was a second revenue stream on a different clock: a network division opened in 1997, spun out in 2004 as Macnica Networks, distributing security products from Silicon Valley vendors such as CrowdStrike into a corporate IT market that semiconductor distributors could not easily enter.

The group extended abroad by acquisition — Hong Kong’s Cytech was consolidated in 2008 — and the two-cycle portfolio kept the damage limited through the financial crisis. By the year to March 2014 Macnica turned over $2.4B (¥256bn), and that portfolio was precisely what made it an attractive partner when talks began with Fuji Electronics, a $446.9M (¥47bn) distributor strong in industrial and automotive accounts. Announced in May 2014, the deal would create Japan’s largest independent semiconductor distributor — roughly 200 suppliers and 10,000 customers, with almost no overlap between the two customer bases.

Macnica Fuji Electronics Holdings was created by joint share transfer on 1 April 2015 and listed the same day. Kamiyama stayed on as honorary chairman; the first president was Nakashima Kiyoshi, who had joined Macnica in 1981 and run the operating company since 2008. In the first year semiconductors were 93% of the ¥405.3bn revenue, and the integration showed up mainly as purchasing leverage — but the small network business earned an 11.5% operating margin against semiconductors’ 1.7%, and reshaping that mix became the stated priority.

Network revenue roughly tripled in five years, from ¥27.2bn in FY2015 to ¥72.8bn in FY2020, as Japanese firms bought endpoint security and then, after 2020, bought it under duress. In June 2019 Hara Kazumasa, forty-eight and a 1995 entrant to the company, took the presidency of both the holding company and the operating company — in a year when the US–China trade friction cut semiconductor revenue by ¥52.1bn and the network business held the group up.

Hara then closed the structure the 2015 deal had left open. On 1 October 2020 the operating company absorbed Fuji Electronics, taking the supplier roster to about 300 and the customer count to about 18,000; on 1 October 2021 it absorbed Macnica Networks as well, ending the unit’s seventeen years as a separate legal entity so that chip engineers and network engineers could design cyber-physical solutions inside one payroll. In August 2022 the holding company dropped the Fuji Electronics name and became Macnica Holdings. Kamiyama stepped back from representative director in 2021 and off the board in 2023.

The global chip shortage of 2021–22 ran straight through Macnica’s books. Semiconductor revenue nearly doubled in two years and its profit contribution rose more than fivefold; group revenue reached $7.3B (¥1.03tn) in the year to March 2023, the first Japanese semiconductor distributor to pass ¥1 trillion, on operating profit of ¥61.6bn. Fifty years after the bakery office, the position Kamiyama had chosen — sole Japanese route for suppliers nobody else wanted — paid at both volume and margin.

Then the industrial inventory correction arrived. Semiconductor operating margin fell from 6.0% in FY2022 to 3.0% in FY2024, wiping ¥30.4bn off that segment’s profit in a single year. The network business went the other way, growing past 20% again to ¥153.9bn revenue and ¥13.3bn profit — about a third of group operating profit, from a segment that is a seventh of group revenue. Macnica cut nothing: headcount rose past 5,000, R&D spending doubled and capital expenditure grew, on the reading that the reaction to a boom is the moment to fund the next structure.

In May 2024 Hara set that reading down as Vision2030 — ¥2 trillion of revenue and ¥150bn of operating profit by FY2030, roughly double the FY2024 base — under the heading of becoming a “service and solution company”: AI and data analytics, managed cyber-security, cyber-physical systems, all of them revenue that does not move with the chip cycle. A three-for-one stock split followed in October 2024, with the shareholder-perk scheme scrapped in favour of dividends, while Kamiyama and his foundation still hold about a fifth of the company. The distributor’s question, after fifty years, is whether it can stop being a distributor.