한국의 반도체 산업: 도전 과제와 전략적 과제

### 반도체 위기: 한국의 과제와 미래 전략
* 📉 애플은 TSMC를 완전히 떠났고, 퀄컴도 떠날 위기에 처해 한국의 반도체 위기가 이어지고 있는 반면, **TSMC는 여전히 강세**를 보이고 **삼성은 고전**하고 있습니다.
* 🇺🇸 미국은 첨단 기술을 거의 독점하다시피 하며 기술 발전에 대한 진정한 의지를 보여 주었고, **이는 패권 국가로의 전략적 전환**을 필요로 합니다.
* 📊 한국 경제는 **64개 첨단 기술 부문에서 중국이 53개 부문에서 선두**를 달리는 등 하락세를 보이고 있습니다.
* 💻 한국은 메모리 반도체 시장에서 **70%의 시장 점유율**을 차지하고 있지만, 마이크론과 같은 경쟁업체들이 반도체 시장을 따라잡으면서 도전에 직면해 있습니다.
* ⚙ TSMC는 **70~80%의 점유율**로 파운드리 시장을 장악하고 있는 반면, 삼성은 그 속도를 따라잡지 못해 **퀄컴과 같은 주요 고객사가 삼성으로부터 주문을 전환**하고 있습니다.
* 💰 반도체 시장은 **2030년까지 1조 달러**에 달할 것으로 예상되며, 필수 기기에 반도체가 내장되면서 현대 생활에서 반도체가 차지하는 중요성이 강조되고 있습니다.
* 🔧 한국은 리더십을 유지하기 위해 **반도체 생산이라는 강점**에 집중하고 노키아와 같은 역사적 사례에서 교훈을 얻어 **첨단 산업 기술에 전략적으로 투자**해야 합니다.
* 🌍 과학기술에 대한 미국 정부의 의지는 **한국이 경제 침체에서 벗어나 선진국으로 도약하기 위해 국가 차원에서 반도체 산업을 지원**해야 할 필요성을 강조합니다.

한국은 글로벌 경쟁력을 유지하고 경제 쇠퇴를 피하기 위해 반도체 산업을 전략적으로 강화해야 합니다.

Apple has completely abandoned TSMC, and Qualcomm is also at risk of leaving, leading to a semiconductor crisis in our country. It seems that TSMC is still running comfortably, but Samsung is struggling to keep up. In fact, the United States is the world's hegemon and has almost monopolized advanced technology. This is not a coincidence; the United States is genuinely committed to technology. Therefore, the strategy of a strong country must completely shift towards hegemonic power. As knowledge increases, we become smarter. You may have heard of the term Peak Out Korea, which suggests that the South Korean economy has now entered a downward trend after reaching its peak. Some people argue that this is a bias against the South Korean economy, but it is indeed the case. If we look at the economic growth rate of South Korea over the past 30 years, we can see the performance of each government and president.

I have been continuing my research on the village of Imgi. I looked into the 64 advanced technology sectors and found that China is already ranked first in 53 of those sectors, while our country does not hold the top position in any of them. What does this mean? It means that the growth engines of our country are dying out. It indicates that new growth engines are not emerging. 

Hello, I am Kim Tae-yu, an emeritus professor at Seoul National University’s Department of Industrial Engineering, and Kim Chang-guk, a partner at the Boston Consulting Group. We are performing well in one part of the immense technology value chain known as semiconductors, but we are not excelling in the whole field. The fundamental technologies are all held by the United States. In terms of materials, components, and equipment, Japan, the United States, and the Netherlands dominate. We then split into foundry and DRAM, and we are only good at producing DRAM. We are facing tremendous challenges and trials.

There is a worrying situation where I wonder when we will lose our dominance in memory. In fact, South Korea is indeed a stronghold in memory production. If we include Samsung and Hynix, our country holds about 70% of the market share. According to a well-known report from IDC, Samsung Electronics ranked first in semiconductor sales, while Hynix ranked third, showing a twofold increase in sales compared to the previous quarter. Additionally, in the case of HBM, Hynix has been announced as having a unique position in the market. 

However, the issue that needs to be raised from here is that in the past, Micron would launch products at least six months to a year later than Samsung and Hynix. But now, Micron has caught up significantly in competitiveness, and recently, they have started to release very similar products almost simultaneously. This indicates that a crisis is emerging for our country's semiconductor industry. There are two areas where our semiconductors are still dominant, as I mentioned earlier.

There are memory semiconductors, and then there is the foundry sector. In the memory semiconductor market, Samsung and SK Hynix together hold a 70% share globally. Samsung Foundry is the only company, after TSMC, that can produce advanced nodes below 14 nanometers. The main issue with memory is how quickly it can be produced according to the customer's needs. For example, when supplying to Apple, if Samsung is ranked first, they might say, You will supply 50% to me. Then, if Hynix comes in second a month later, they might say, You will supply 30%. If Micron comes in third after six months, they would then supply 20%. This means that Micron's supply volume is relatively small. Additionally, when semiconductors are initially mass-produced, the yield is usually around 80%. However, Micron is coming in six months later.

Samsung and Hynix already have a yield of 95%, while Micron had 80%. This means that Micron's costs are higher due to their lower yield. However, they have been catching up quickly, and now the pace of advancement is slowing for us while it is accelerating for them. Micron was the first to release the 8-high stack 24 GB HBM3 into the market in July of last year. As a result, due to the effect of competitors catching up, we are facing a crisis. It has become more difficult for us to make money and maintain our top position as we did in the past. This is the first issue. The second issue is foundry. What other semiconductors does our country excel in? Within foundries, only two companies can produce chips at a node smaller than 14 nanometers: TSMC and Samsung. So, there are only two companies, but the problem lies in market share. TSMC holds 70% to 80% of the market.

Recently, you may have seen several news reports about our country, particularly regarding the 20% share of the semiconductor market. There have been discussions about Samsung's failure with the 3-nanometer process and how TSMC has gained a significant advantage. TSMC continuously advances to the next generation, releasing new products every year. They had 5-nanometer technology two years ago, 4-nanometer last year, and now 3-nanometer this year. This rapid innovation, advancing one step each year, is remarkable. However, it seems that TSMC is still moving forward comfortably, while Samsung is struggling to keep pace. For instance, last year, Qualcomm announced that they would stop allocating orders to Samsung Foundry this year. The reason for this decision was that the quality expected from the nanometer process was not being met, prompting Qualcomm to turn to TSMC instead. Originally, both Apple and Qualcomm relied on Samsung, but now they are shifting towards TSMC.

For a year, we deliberately engaged in competition by alternating years, but now Apple has completely gone with TSMC, and Qualcomm is on the brink of a departure. This industry now requires us to continuously invest 20 to 30 trillion won each year, step by step, and it is becoming increasingly difficult. So, is there no way for us to protect our interests while narrowing the gap with developed countries? This is indeed a very important issue. The solution lies in what we excel at, which is semiconductors. We should focus on semiconductors. I argue that if we concentrate on what we do well, we will become even better at it. There is a historical example: the Netherlands is a very special country. Its land area is just 40,000 square kilometers, which is only about one-fiftieth of the Korean Peninsula, making it a very small country. Yet, at one time, the Netherlands became one of the three global powers. What happened?

After that, the North Sea herring fishing grounds were formed. So, to catch herring, everyone from England, the Netherlands, Germany, and France rushed in. However, the Netherlands completely triumphed. Why? Because while fishermen from other countries would catch herring, sell it, and then catch more to sell again, the Netherlands essentially focused their entire society on herring fishing. They created dedicated herring fishing boats and sent them out to catch herring. As soon as they caught the fish, they salted them. They also brought in fast boats. At one point, more than half of the Dutch population was involved in industries related to herring fishing, making the Netherlands a true herring fishing nation. So, how does catching herring well lead to becoming a developed country? The truly important fact here is that catching herring...

In order to catch herring, they established a fishing ground and created a fast vessel. It turned out that the shipbuilding industry was rapidly developing, and they became a great shipbuilding nation, producing excellent ships. Because they made good ships, they were also able to produce trade vessels very well. When we consider a ship weighing about 250 tons, the British trade ship required around 30 crew members to operate. However, the trade ship created by the Dutch, known as the fluyt, only needed about 10 crew members. This reduced labor costs to a third. As a result, the British had no way to compete with the Dutch. The Dutch ships were more efficient, so they started selling ships and directly engaging in shipping. They ventured into shipping and monopolized the trade of spices, such as pepper, cloves, and nutmeg, which at the time were referred to as black gold. So what this means is...

If we create a significant gap by excelling in one area, new industries will emerge from that gap. This is what we call technology spillover; technology keeps overflowing and leads to the creation of new industries. Strong nations have large territories and populations, which means they have many researchers and a large consumer base once products are developed. Therefore, they need to focus on balanced development of technology. However, our country is a small but strong nation. This means we have fewer researchers and consumers, so we can't excel in producing every product. Even if we create something, the market isn't large enough. Thus, rather than trying to produce products we are not good at, we should focus on improving the products we can make well and find breakthroughs from there. The strategies of strong nations and small but strong nations are different, and so our strategy as a small but strong nation is...

We need to become better at what we do. If we invest more and focus more, we can definitely succeed. Currently, the total revenue of the semiconductor market is about 600 billion dollars. Experts predict that by 2030, it will reach one trillion dollars. This is a business that always experiences double-digit growth. In the past, we used to refer to the basic necessities of life as food, clothing, and shelter, but now I call it food, clothing, and devices. The term devices signifies that these have become essential items in our lives. For instance, whether we are driving a car or flying on an airplane, we rely on various devices. What we often overlook is that a significant amount of semiconductors is embedded in these devices. In particular, the amount of semiconductors in a car can range from 400 to 600 dollars.

The cost of semiconductors is about 4% to 6% based on the cost price. This means that semiconductors are now considered essential goods in our daily lives, just like food, clothing, and shelter. Another point is that when I produce something, it means I am self-sufficient. The importance of self-sufficiency has grown significantly. Many of you may have experienced this: in the past, when you wanted to buy a car, you could get it right away. However, during the COVID-19 pandemic, cars suddenly stopped being produced for a year. This was due to a lack of semiconductors. During the pandemic, it became clear that without semiconductor production in our country, I wouldn't be able to buy what I really needed when I needed it. If our country doesn't sell them, we can't produce them, which was a challenging time. This situation created a fear that countries could use semiconductors to exert pressure on each other. If we lose semiconductor dominance, we will fall behind in various industries.

The concern is that we will either follow or become dependent on imports, which is problematic. This is why we see the emergence of semiconductor production facilities and friend-shoring initiatives. Strategically speaking, even though we are good at semiconductor production, we cannot maintain a monopoly or hegemonic position on our own; we must work together for mutual growth. Even countries like the United States need to collaborate for co-prosperity. 

So, to avoid losing our leadership, what kind of strategy should we devise? It's a tough question. Ideally, we should do both. However, given the current situation, the U.S. is making efforts to bring in chips that are already being produced. By doing so, they believe they can control something in their territory if issues arise later. The first point is...

The second important topic is about talent. It ultimately comes down to the expertise that people possess, including the know-how related to semiconductor production. For example, if a memory plant is established in the U.S., and if the people there stay for a long time, the talent pool will eventually shift to other companies within the U.S. This creates a strength in continuously enhancing their proprietary know-how. On the other hand, if I say that I will nurture a semiconductor company in this country, the current reality is that 90% of new semiconductor startups are foundries. Why? Because there are Qualcomm, NVIDIA, and many others. In fact, aside from the U.S. foundries, there are only a couple in Taiwan that can generate over $3 billion in revenue. Most startups cannot initiate production in important sectors. Therefore, no matter how strong the incentives the government provides,

Even if we provide support, it takes too long for such companies to grow and contribute to the semiconductor ecosystem of their country. Therefore, when attracting investments that can produce larger quantities, all countries are focusing heavily on direct incentives. We are also implementing the K-Chips Act, but we don't offer direct incentives for investments, which I always argue is the least effective approach. We are currently moving towards a corporate tax-centered strategy. Why is the effect of corporate tax weak? Because corporate tax is paid after generating sales and profits. However, our semiconductor production needs to be advanced in preparation for next year. Since corporate tax comes into play afterward, it holds very little significance. Regarding today's topic, we have received questions in advance from our subscribers about how national strategy is involved in fostering industries, which ultimately requires government financial input, particularly for semiconductor companies.

Although it is a large corporation, should the government support companies? The importance of advanced industrial technology does not lie in helping specific private enterprises. Countries that genuinely invested in advanced industrial technology have all succeeded and become advanced powers, while those that did not have faced significant difficulties. For example, there was once a company called Nokia in Finland. Nokia accounted for 25% of all exports from Finland and 70% of the market capitalization of Helsinki. At that time, people said that if Nokia failed, Finland would fail too. But Nokia did fail. However, it is said that the fallen Nokia revived Finland. Nowadays, Finland has 25% of its startups coming from the technology developed at Nokia, despite its population being only 4% of the EU.

People who studied and worked at Nokia researched other technologies that were not adopted there and then ventured out. This means that if we invest heavily in advanced industrial technology, there can be difficulties if we over-invest beyond the country's capacity. Still, technology does not disappear; it can eventually lead to success. Nokia serves as a case in point. So where does all that money come from? We are betting stronger than the United States, but we are a smaller country with fewer resources. How do we manage that? I believe we must do it, even if it's challenging. This is not a business where large corporations make money. There is a concept called gross value added. This pertains to how much the entire nation contributes to added value, with the first being the national added value obtained from semiconductor production.

Exports are approximately around 100 billion dollars. The second aspect is related to semiconductor production, which includes materials and components, generating about 90 million dollars. Lastly, there are the ripple effects; for example, in our country, companies like Samsung and SK Hynix lead to the development of surrounding complexes and a significant increase in apartment constructions, as well as activated economic activities. All of this adds up to about 100 billion plus additional factors, resulting in roughly 400 trillion won in our currency. Therefore, the economic value is incredibly large, with an economic added value of 400 trillion won. This is why everyone has started to focus on semiconductors. Ultimately, we need to consider whether this should be viewed as a business where large corporations make money or within the scope of national industrial development, taking into account the ecosystem and the subsequent ripple effects.

I believe that we should focus on what we need to aim for. In fact, I think the latter perspective is correct. The United States is currently the world's hegemonic power, and it monopolizes almost all advanced technologies. This is not a coincidence. The U.S. is a country that is genuinely committed to technology. It has begun to pour enormous national budgets into science and technology. Through institutions like the National Science Foundation, a significant amount of national funding has been sent to universities, resulting in the emergence of many young prodigies. Silicon Valley, which has become the center of global technology, is built around these prodigies. Therefore, today's U.S. hegemony was not achieved by chance; it is because the U.S. government and its people truly understand how crucial industrial technology, especially advanced industrial technology, is.

So now, in order for our country to break free from the peak-out of economic growth rates and truly rise as a developed nation, what we need to do is to strategically support advanced industrial technology at the national level. Today, I would like to emphasize that you shouldn't worry about calorie intake; instead, try the keto diet. Yes, download the SBS Premium app now.

[지식뉴스] "한국, 더 이상 1등 어려워"...'위기감 최고조' 한국 반도체 생존전략② (ft.김태유 서울대 산업공학과 명예교수) / 교양이를 부탁해 / SBS

지식플랫폼 스프 앱 출시 지금 다운로드!

비디오머그 - VIDEOMUG

애플이 TSMC를 포기하고 퀄컴도 떠날 위기에 처하면서 국내 반도체 산업에 위기가 닥쳤습니다.

미국은 첨단 기술을 독점하고 있는 반면, 한국 경제는 하향세에 접어들고 있습니다.

메모리 생산량에서 세계 시장 점유율 70%를 차지하고 있지만 마이크론의 경쟁력 추격으로 한국의 지배력이 도전받고 있습니다.

삼성과 하이닉스의 수율은 95%인 반면, 마이크론은 80%로 마이크론의 비용이 더 높지만 빠르게 따라잡고 있습니다.

TSMC와 삼성은 14나노미터보다 작은 노드에서 칩을 생산하는 유일한 두 회사이며, TSMC는 시장 점유율의 70~80%를 차지하고 있습니다.

발표자는 선진국과의 격차를 좁히기 위해 네덜란드의 청어 어업 성공 사례를 들며 반도체에 집중해야 한다고 주장합니다.

네덜란드 조선업이 무역과 기술에 미치는 영향

네덜란드는 어장과 빠른 선박을 확보하여 조선 산업과 효율적인 무역선을 빠르게 발전시켰습니다.

네덜란드의 무역선인 플루이트호는 더 적은 수의 선원이 필요해 인건비를 절감하고 경쟁 우위를 확보할 수 있었습니다.

이러한 효율성 덕분에 네덜란드는 향신료 무역을 독점할 수 있었고, 기술 파급과 산업 발전을 위해 우수한 분야에 집중하는 것이 중요하다는 점을 강조했습니다.

반도체는 현재 전체 비용의 4~6%를 차지하며 필수품으로 간주되어 자급자족과 생산에 영향을 미칩니다.

코로나19 팬데믹으로 인해 반도체 생산에 대한 의존도가 부각되면서 국가 간 힘의 역학 관계에 대한 우려가 커졌습니다.

반도체 생산에서 리더십을 유지하려면 협업과 인재 육성이 중요하지만, 대부분의 신생 스타트업은 파운드리 업체로 주요 부문에서 생산을 시작하기가 어렵습니다.

반도체 산업 육성을 위한 정부의 역할

각국은 반도체 생산을 위한 투자를 유치하기 위해 직접적인 인센티브에 집중하고 있지만, 한국형 반도체법은 법인세 중심의 접근 방식을 취하고 있습니다.

법인세는 매출과 이익이 발생한 후에 납부하기 때문에 실효성이 약하고, 첨단 반도체 생산에 대한 시급한 수요와도 맞지 않습니다.

노키아가 실패한 후 핀란드의 스타트업 생태계에 미친 영향에서 알 수 있듯이 첨단 산업 기술에 대한 투자는 장기적인 성공으로 이어질 수 있습니다.

반도체 관련 수출액은 약 1,000억 달러에 달하며, 생산은 9,000만 달러를 추가로 창출합니다.

삼성과 SK하이닉스와 같은 기업의 입주로 인해 아파트 건설 증가, 경제 활동 활성화 등 약 400조 원에 달하는 경제적 파급 효과가 발생했습니다.

미국 정부의 기술 및 전략적 자금 지원은 국가 발전을 위한 첨단 산업 기술 지원의 중요성을 강조하며 미국 정부의 글로벌 지배력에 결정적인 역할을 했습니다.