Article count:25311 Read by:103629709

Samsung Semiconductor's Manufacturing Ambitions

Latest update time：2021-10-28

Reads：

Source: The content comes from Semiconductor Industry Observer compiled from Nikkei and others, thank you.

Samsung Electronics said on Thursday it plans to strengthen its foundry chip technology and add customers as a global chip shortage disrupts production in key industries from automobiles to smartphones, which also helped them post strong third-quarter earnings, Nikkei reported.

The South Korean tech giant is boosting the competitiveness of its foundry business — which makes tailor-made chips for corporate clients — with plans to make those components more cost-effective. The company announced earlier this month that it plans to produce its first 3-nanometer chip designs for customers in the first half of 2022, while its second-generation 3-nanometer chips are expected to be produced in 2023.

"The foundry business is expected to continue to achieve strong performance improvements by securing technology leadership through the 3nm all-gate process and meeting demand through aggressive investments," Samsung said in a statement after announcing full third-quarter earnings.

The company plans to significantly improve earnings by expanding supply of advanced processes below 5nm and increasing sales to global customers, as well as normalizing prices for future sustainable investments,” they added.

The company said net profit for the July-September period surged 31.3% from the same period last year to 12.3 trillion won ($10.5 billion).

Samsung also said third-quarter operating profit rose 28% year-on-year to 15.8 trillion won, in line with guidance issued earlier this month. But revenue was revised upward, rising 10.5% to 74 trillion won over the same period. Month-on-month, operating profit rose 25.9% and revenue rose 16.2%.

Analysts say Samsung's foundry business will triple in five years, adding hundreds of customers.

“Samsung Foundry is providing a stable platform and cooperation ecosystem to customers,” Lee Su-bin, an analyst at Daishin Securities, wrote in a note ahead of the earnings release.

Lee said Samsung's customer base is increasing dramatically, from 35 in 2017 to more than 100 this year. Lee expects it to exceed 300 by 2026.

By sector, memory chips led gains due to strong demand for DRAM and NAND chips during the coronavirus pandemic. The mobile segment also contributed as Samsung's latest foldable smartphones - the Galaxy Z Fold3 and Flip3 launched in August - gained popularity around the world.

Despite strong earnings this year, Samsung's shares have underperformed as investors worry that a cyclical downturn in memory chips could soon set in and weigh on the company's value.

Samsung shares fell 0.29% in early trading after the company released its earnings report before the market opened.

The Economist: Samsung faces many difficulties in dominating chip manufacturing

Samsung Electronics is a behemoth. The South Korean tech company is the crown jewel of the powerful Samsung chaebol. It makes more smartphones, as well as more home entertainment systems and appliances, than any other company in the world. It also dominates the manufacture of memory chips, which are used to store data on electronic devices and whose prices have been driven up by a global semiconductor shortage. Samsung's $200 billion in annual revenues aren't much less than Apple, the company with the highest market value in history, and it's sitting on a $100 billion cash pile.

Now, Samsung and its parent company are entering a critical new chapter. In August, Lee Jae-yong, a descendant of the founder who founded Samsung in 1938, was released from prison after serving two stints in jail following a conviction for his role in a bribery scandal. He then took full control of the empire from his father, Lee Kun-hee, who died last year.

Now free, Mr. Lee has big plans for the company, which he wants to dominate in producing cutting-edge logic chips used to process information, as in memory and smartphones. That would put them head-to-head with chipmaking giants like Taiwan's TSMC and the United States' Intel, and thrust them into a fierce global competition for control of one of the world's most strategic industries.

On October 7, the company confirmed that it will make some of the world’s most advanced logic microprocessors in 2022, based on its new “GAA” architecture with three-nanometer (billionths of a meter) transistors. It also surprised analysts by announcing plans to mass-produce two-nanometer chips starting in 2025. It expects to invest around $37 billion in capital expenditures across its business this year. It is winning new customers, such as U.S. chip designer Nvidia and electric car maker Tesla.

The outcome of Mr. Lee’s gamble will have far-reaching consequences — and not just for Samsung. It’s important to South Korea that its president considers Mr. Lee’s parole to be in the national interest, given the chaebol’s importance to the economy. It will affect the global semiconductor industry, whose critical nature has been highlighted by a global chip shortage. To ensure success, a man described by acquaintances as shy, respectable and shrewd must also summon a degree of ruthlessness.

Samsung is a complex business with strategic challenges and a poor stock market performance. It is best understood as being divided into two main businesses. The first makes "sets": smartphones, TVs and home appliances. The second produces "components" that go into Samsung's own devices and are sold to external customers such as Apple. Samsung further split its TV business into two divisions: appliances such as TVs and washing machines, and digital devices (primarily smartphones). The components business, meanwhile, includes semiconductors and displays.

But those businesses won’t be Samsung’s future growth engines. People close to the company say that in Mr Lee’s hierarchy of Samsung businesses, home appliances sit at the bottom, below the TV division, with similarly low margins but a greater role in strengthening Samsung’s valuable brand. Next up is the phone business, which in the early 2010s contributed more than half of profits. It continues to generate plenty of cash despite repeated concerns about its future, and there’s some new optimism as new phones with foldable screens sell well.

At the top of the hierarchy are semiconductors. Historically, Samsung has focused on memory chips, where it has a 44% global market share in DRAM chips (temporary storage for desktops) and NAND devices (permanent storage for mobile devices). The memory business brings in a little over 20% of revenues but close to half of operating profits (see chart below). Everything else is likely in the service of high profits.

Insiders said that if there is a disagreement between a lower-tier business and the component division on pricing or other terms, the component business takes precedence. According to the company, its unique ecosystem benefits from diversified businesses that allow internal innovation while providing stability during the ups and downs of industry cycles.

Changing storage market

Analysts believe Samsung’s memory business has plenty of room to run. Nicolas Gaudois of UBS bank said that because such chips are critical for storing data across industries, it “can only go in one direction: that’s up”. Research firm Omdia predicts that the global memory chip market will grow by double digits a year from 2020 to 2025. It is now less cyclical due to surging demand from data centers and, on the supply side, market consolidation. Increasingly extreme miniaturization means competitors can no longer improve production efficiency as easily as before. Samsung says it has proven its ability to innovate and extract value in mature businesses. Internally, though, some senior executives worry that memory is a mature operation. Some investors worry that demand for memory chips may weaken by the end of the year.

One option is to follow Apple in developing its services business, which has grown to a fifth of the iPhone maker’s revenue from 8% in 2012. Yet despite some successes, particularly in payments and health apps, Samsung’s efforts to add software and services to its world-class hardware have been sporadic.

That’s partly because Samsung’s hardware-first approach is deeply embedded in its culture. It’s likely reinforced by Lee’s personality and experience. “He’s very cautious and conservative by nature, even more so than his father,” one former executive said. That innate conservatism may have been reinforced by his first major endeavor after attending Harvard Business School. In the late 1990s, at the height of the dot-com bubble, he invested in the venture capital firm eSamsung. Watching the subsequent bankruptcy, the former executive said, made Lee skeptical of Korean software engineers. eSamsung was subsequently shut down, too.

A big push into services could also jeopardize Samsung’s long-standing successful partnerships with software giants like Google and Microsoft. In 2014, Samsung launched a music streaming service called Milk Music, but despite its success, it was canceled two years later. “Google saw Samsung’s software work as fragmenting the Android ecosystem and felt threatened,” one former executive recalled. “I’m pretty sure Samsung has given up on software and services,” he sighed. He’s worried about missing out. Even if the company talks about trying again, it’s probably just to keep Google and other partners honest, he added.

Another issue is China. The country is a significant source of demand for memory and logic chips. To meet its needs, Samsung will complete its second memory chip factory in the western city of Xi'an this year. Despite growing tensions between China and the West, especially the United States, neither domestic nor any other Korean chipmakers are likely to abandon their giant neighbor, which will likely remain a big buyer for many years (especially for more technologically complex chipmakers). This means Samsung must walk a careful line to retain Chinese customers while not abandoning American ones.

This array of complications and risks helps explain Samsung’s underperformance relative to other giants, as it faces stiff competition in both consumer technology (China’s Apple and Xiaomi) and chipmaking (TSMC and Intel). The company also suffers from a group discount. It is listed only in Seoul, and restrictions on exposure to individual stocks in the past have prompted local investors to sell Samsung when its shares soared. Samsung accounts for nearly a fifth of the Kospi stock market index. And Samsung’s huge cash pile has weighed down returns.

As a result, despite solid operating results, Samsung’s shares still trade at one to one and a half times forward book value, well below peers. Raising its dividend from 22% of net profit in 2018 to 78% in 2020 helped Samsung’s market value more than double in the two years to January. But Apple’s has almost tripled over the same period. The strong outlook for semiconductors and the less cyclical nature of memory chips have yet to translate into higher valuations. Samsung’s market value surged nearly 50% in late 2020 but is down 13% since the start of the year, while New York’s tech-heavy Nasdaq and a basket of global chipmakers have both risen (see chart).

Logic chips are on the move

Mr. Lee’s bet on cutting-edge logic chips is aimed at turning around poor performance. The idea is to win a big piece of the fast-growing and lucrative non-memory chip market, which accounts for 70% of the $550 billion global semiconductor market. Mr. Lee has set a goal of matching Samsung’s roughly 40% market share in memory in the “foundry” business of making processors for customers.

Samsung's successor has his work cut out for him. Samsung's foundry division took a hit in 2016 when Apple moved all business for the iPhone's A-series processors to TSMC. The shock provided a stark example of how Samsung's complex structure can unravel potential conflicts of interest with major customers. Half of Samsung's foundry capacity goes to other divisions of the company's devices, with the rest supplied to external customers. Apple prefers pure-play foundry TSMC to Samsung because it competes with it in the smartphone space.

So far, progress has been slow on the ambitious goals first outlined by Mr Lee a few years ago. The company has about 15% of the foundry market, compared with more than 50% for TSMC, which plans to spend $100bn on new capacity over the next three years. Samsung's non-memory chip revenue accounts for just 7% of total sales (although that was from scratch in 2005, and the company also makes some other specialized processors for sensors and the like). If you count profit share, it's even lower.

Perceived conflicts of interest aren’t its only challenge. While the memory and logic businesses share some commonalities and overhead, they differ in important ways. Producing memory chips is mostly about speed, volume and economies of scale. Making high-end logic processors is far more technically complex, with engineering done at the nanometer level and customers increasingly wanting custom chips for their purposes.

In technology, it’s fair to say that almost everyone else has lagged behind TSMC in at least the last two generations of cutting-edge processors. Part of the reason may be sensible caution. But a semiconductor executive at another company said the reticence would further complicate relations with customers, many of whom are reluctant to place orders unless they can get guarantees of capacity. Rather than anticipating their needs, Samsung is reacting, the executive said.

Aware of these problems, Lee clearly wants to accelerate Samsung's transformation. The company is leveraging its research and development (R&D) strength and taking some risks on next-generation logic chips, such as adopting its new advanced chip architecture. The company does not disclose how much of its capital expenditure is for memory chips and how much for logic. According to brokerage CLSA, the focus is on logic chips, which are also R&D-intensive.

The expanding constellation

Samsung is also considering building a $17 billion plant in Texas to make cutting-edge logic chips to appease U.S. desire to bring more chip manufacturing back home from Asia. Sanjeev Rana of CLSA noted that the new customers it is courting, such as Nvidia and Tesla, do not overlap with its other businesses.

The semiconductor industry's fraught geopolitics could help. While rising tech nationalism in chip design and manufacturing has governments favoring domestic production and local champions, it could ultimately benefit them. Meanwhile, concerns about TSMC's future are growing. One semiconductor executive said many companies that use TSMC are scrambling to reduce their dependence on the Taiwanese company as a precaution. Samsung, TSMC's closest competitor, could be a big beneficiary. Samsung has the world's largest industrial complex of semiconductor factories and engineers, as well as some of the best chip technology, said Mark Newman, a former Samsung Group executive and chief business officer of battery startup Nyobolt.

One way to speed up the transformation is to split it into its component businesses, as investment bankers have long suggested. That would also remove potential conflicts of interest that have hobbled Samsung’s foundry unit. Meanwhile, a dual listing in the U.S. could help address Kospi-related drags.

Yet neither a breakup nor a relisting looks likely. Mr Lee seems unwilling to back the radical first option. An attempt to convince them to enter a second one, as part of an activist campaign by US hedge fund Elliott Management around 2016, failed. Realising this, shareholders are therefore pressuring Samsung to at least do something with its unused cash. One idea is to pay them 100% of free cash flow. Alternatively, Samsung could make a big acquisition. The company said that "the founding family is clearly aligned with all other shareholders in their goal of creating maximum value and having that value properly reflected in the market."

To have a significant impact on Samsung's financial results, any deal would have to be big. Lee's inclinations and preferences make such a gamble on software and services impossible. That leaves chipmaking as where the company's cash could be spent. One potential acquisition target is NXP Semiconductors, a Dutch company that specializes in the fast-growing automotive chip market. At a market cap of $50 billion, that would be a tall order, but not impossible.

If Samsung Electronics is to become a logic chip star to compete with TSMC, Mr. Lee had better get promoted. Last year, he vowed not to hand over management of Samsung to his children (although Lee will likely retain the largest stake in the company through various family-controlled businesses). Mr. Lee’s pledge to be the last person to run the company, along with other improvements in corporate governance that insiders say have cleared a path to the top for its many talented executives. They must hope that Mr. Lee’s legacy to them is less complicated than his father’s.

Original link:

https://asia.nikkei.com/Business/Technology/Samsung-strengthens-foundry-business-amid-global-chip-shortage

https://www.economist.com/business/2021/10/21/samsung-electronics-wants-to-dominate-cutting-edge-chipmaking

*Disclaimer: This article is originally written by the author. The content of the article is the author's personal opinion. Semiconductor Industry Observer reprints it only to convey a different point of view. It does not mean that Semiconductor Industry Observer agrees or supports this point of view. If you have any objections, please contact Semiconductor Industry Observer.

Today is the 2841st content shared by "Semiconductor Industry Observer" for you, welcome to follow.

Latest articles about

■TSMC's 2nm is too powerful, UMC is too miserable

■The United States has heavily funded this semiconductor technology

■Tesla is also snapping up HBM 4

■Intel's next-generation AI chip is exposed for the first time

■Nvidia releases its largest chip yet

■The Danish robot giant invites you to "do something" together

■Self-developed DPU released: Microsoft chip, full of firepower

■In the post-Moore era, optical computing chips have become the key to breakthrough, and domestic manufacturers have great potential!

■Open source software in crisis

■The global semiconductor equipment giants are all in trouble