Hello, I am Ken Hsiang, the Head of Investor Relations for ASE Technology Holding. Welcome to our second quarter 2022 earnings release. Thank you for attending our conference call today. Please refer to our safe harbor notice on page two. All participants consent to having their voices and questions broadcast via participation in this event. If participants do not consent, please disconnect at this time. I would like to remind everyone that the presentation that follows may contain forward-looking statements. These forward-looking statements are subject to a high degree of risk, and our actual results may differ materially. For the purposes of this presentation, our dollar figures are generally stated in New Taiwan dollars, unless otherwise indicated. As a Taiwan-based company, our financials are presented in accordance with Taiwan IFRS.
Results presented using Taiwan IFRS may differ materially from results using other accounting standards, including those presented by our subsidiary using Chinese GAAP. Intercompany transactions between our ATM and EMS businesses have been eliminated during consolidation. For today's call, I am joined by Dr. Tien Wu, our COO, and Joseph Tung, our CFO. During the call, Dr. Wu will first provide a mid-year update and overall industry outlook. I will quickly go over our financial results, and Joseph will provide the third quarter outlook. Tien and Joseph will both be available to answer questions during the Q&A session that follows. Also, as a reminder, we disposed of ASE, Inc.'s China sites at the end of 2021.
For our financial results presented here, in addition to our legal entity results, we have included information on a pro forma basis, or as if the disposition of ASE, Inc.'s China sites had already occurred. We believe the pro forma results give additional meaningful information, which would assist in providing comparability of our financial results. For the purposes of this presentation, including that of Dr. Wu's, we will generally discuss our full company and ATM second quarter results sequentially, compared with first quarter legal entity results and year- over- year compared with pro forma second quarter 2021. Dr. Tien Wu will now present our mid-year update. Dr. Wu?
Good afternoon. I would like to give you some highlight. First, I would like to talk about our first half 2022 and second quarter performance. Our second quarter ASE Hold Co revenue grew 27% year-over-year in US dollar term. First half 2022 revenues grew 28% year-over-year. As Ken already pointed out, all on the pro forma basis. Second quarter 2022 ASE ATM revenues grew 25% year-over-year, while the first half 2022 revenue grew 25% year-over-year. First half 2022, we saw Advanced Packaging revenue up 48% year-over-year. First half 2022 ASE Hold Co automotive revenue grew 64% year-over-year, while the ATM automotive revenue in the first half grew 54% year-over-year.
We do expect the momentum to continue into the second half, as well as 2023 and 2024. First half 2022 Hold Co operating margin improved 2.3 percentage points, out of which 0.5 percentage points were from the favorable currency. First half 2022 ATM operating margin increased 3.4 percentage points, out of which 0.6 percentage points were from the currency. Let me turn to the next page. I would like to give you a highlight of the 2022 full year. First of all, our full year outlook is on track. The overall market is undergoing inventory correction, with some sectors more aggressive than the others. However, we're still seeing some sectors remain constrained. With our diversified customer portfolio and manufacturing flexibility, we're seeing a solid second half 2022 with quarter-over-quarter growth of our Hold Co revenue.
2022 full year ATM revenue year-over-year growth will be 2x of the logic semiconductor industry, with EMS also seeing solid top-line growth. We do expect further margin expansion for both ATM and EMS business comparing to 2021. The next page, I would like to give you some highlight for the accomplishment for the past few years. Namely, I would like to give you a structural improvement of our efficiency and margin. For the past two years, we have made great strides in the ASE and SPIL synergy in R&D, operation, capacity planning, business consolidation, and also the customer portfolio calibration and procurement. With that, the synergy is offering us some percentage of margin improvement. We also made a lot of effort in the automation. The automation has improved significantly our ability to entertain high volume, high reliability business.
Has also improved our manufacturing efficiency, cost structure, and flexibility. We do see increasing demand in multi-die co-packaging. That trend is adding the complexity of ATM know-how, and therefore, we believe ASE's value in the total supply chain at a system level. With the above three, the synergy, automation, and also the value, we do believe that we have a structure improvement of our margin structure in mid- single- digit. That is, going forward, we will see higher peaks and shallower troughs in future cycles comparing to our past performance. Now, the last page, I would like to give you some business outlook. For next year, I know many of you are concerned about the business state. It is too early to comment on the macro environment and potential end market demand shift. However, the general perception is seasonality will resume in Q1 2023.
We believe backend capacity is in a healthy position. Because if you look at the overall investment for the past few years in backend, the incremental backend capacity added is relatively low when compared with the front end. Especially considering backend investment are needed for technology migration as well as volume expansion. When I talk about technology migration, I was implying to density, package type, as well as multi-die. That is, for the same equipment in the advanced technology, you might only be able to support fewer units due to the complexity and density. We are also seeing increasing resource allocation or resource demand needed for NPI after more than two years of COVID lockdown. The NPI trend is very healthy.
We're monitoring the NPI trend for all customers because that is a key indicator which can set our direction for future capacity investment, planning, equipment type, upgrade, automation, and technology roadmap. Lastly, we are optimistic that ASE's capacity utilization will stay at a higher level given our customer engagement, LTA, technology, automation leadership, manufacturing scale, and as I have pointed out in the last few bullets, relevance to all NPIs. Thank you. With that, I'll pass to Ken.
Thank you, Tien. Let's quickly go over the second quarter financial results. Please turn to page seven, where you will find our second quarter consolidated results with legal entity and pro forma basis comparisons. For the second quarter, we recorded fully diluted EPS of TWD 3.61 and basic EPS of TWD 3.69. Consolidated net revenue increased 11% sequentially and 33% year-over-year. We had a gross profit of TWD 34.4 billion with a gross margin of 21.4%. Our gross margin increased by 1.7 percentage points sequentially and 2 percentage points year-over-year. The sequential margin increase is primarily attributable to favorable currency conditions within our ATM and EMS businesses.
From an annual perspective, the margin improvements are primarily the result of higher profitability, scale efficiency, and a favorable currency environment within our ATM business and scale efficiencies within our EMS business. Our operating expenses increased sequentially by TWD 1.4 billion during the second quarter to TWD 13.8 billion, primarily as a result of higher bonus and profit sharing expenses during the quarter. On a year-over-year basis, our operating expenses increased by TWD 2.7 billion, mainly from the increase of scale in both our ATM and EMS businesses. Our operating expense percentage stayed flat sequentially at 8.6%. On an annual basis, our operating expense percentage declined 0.6 percentage points from 9.2%. Improvements in operating expense percentage were achieved as a result of operating leverage created.
Operating profit of TWD 20.6 billion, up TWD 4.5 billion sequentially and TWD 8.2 billion year- over- year. Operating margin was 12.8%, increasing 1.6 percentage points sequentially. Operating margin increased 2.5 percentage points on an annual basis as a result of higher loading and increased profitability. During the quarter, we had a net non-operating gain of TWD 0.5 billion. The non-operating gain was primarily from our net foreign exchange hedging activities, offset in part by net interest expense of TWD 0.7 billion. Tax expense for the quarter was TWD 4.5 billion. The effective tax rate for the second quarter was 21.2%. During the quarter, we saw slightly higher tax expenses, primarily related to undistributed earnings and treasury activities.
We expect a full year effective tax rate of between 20%-21%. Net income for the quarter was TWD 16 billion, representing an improvement of TWD 3.1 billion sequentially and TWD 5.7 billion year-over-year. The US dollar strengthened against the NT dollar and the Chinese yuan during the second quarter. Sequentially, we estimate that currency fluctuation had a 1.6 percentage point beneficial impact to our holding company gross margin. From a year-over-year perspective, we estimate that currency fluctuation had a 1.4 percentage point positive impact to gross margin. On the bottom of the page, we provide key P&L line items without the inclusion of PPA-related expenses. Consolidated gross profit excluding PPA expenses would be TWD 35.3 billion with a 22% gross margin.
Operating profit would be TWD 21.8 billion with an operating margin of 13.6%. Net profit would be TWD 17.2 billion with a net margin of 10.7%. Basic EPS excluding PPA expenses would be TWD 3.97. On page eight is a graphical presentation of our consolidated financial performance. On page nine is our ATM P&L with historical results on a legal entity and pro forma basis. It is worth noting here that the ATM revenue reported here contains revenue eliminated at the holding company level related to intercompany transactions between our ATM and EMS businesses. During the second quarter, our ATM business ramped up significantly ahead of where we thought it would. The revenue level achieved in the second quarter is near our original estimation of our third quarter revenues.
It almost goes without saying that capacities were tight with overall demand for our services remaining strong during the quarter. From the cost perspective, and as we mentioned last quarter, we encountered some higher costs of operations during the quarter, namely logistics, labor scarcity, lower efficiencies from COVID, and higher energy costs. These incremental costs offset the scale efficiencies that were created during the quarter. Going forward, while costs related to COVID and labor shortage issues appear to be more under control, energy costs appear to be more ongoing. For the second quarter, revenues for our ATM business were a record TWD 95 billion, up TWD 11 billion from the previous quarter and up TWD 22.3 billion from the same period last year. This represents a 13% increase sequentially and a 31% increase year-over-year.
Our ATM revenues came in ahead of our expectations due to broad-based higher than expected loading. Gross profit for our ATM business was TWD 27.8 billion, up TWD 4.7 billion sequentially and up TWD 8.9 billion year-over-year. Gross profit margin for our ATM business was 29.2%, up 1.7 percentage points sequentially and up 3.3 percentage points year-over-year. The sequential gross margin improvement was primarily due to NT dollar depreciation. The year-over-year gross profit margin improvement was primarily attributable to NT dollar depreciation and scale efficiencies. During the second quarter, operating expenses were TWD 9.8 billion, up TWD 0.7 billion sequentially and TWD 2 billion year-over-year.
Our operating expense percentage was 10.3%, down 0.5 percentage points sequentially and year-over-year. During the second quarter, operating profit was TWD 18 billion, representing an increase of TWD 4 billion quarter-over-quarter and an improvement of TWD 7 billion year-over-year. Operating margin was 18.9%, improving 2.2 percentage points sequentially and 3.7 percentage points year-over-year. The NT dollar depreciating against the US dollar had a positive 2.3 percentage point impact on our ATM sequential margins. On a year-over-year basis, we estimate that the strengthening US dollar had a 2 percentage point positive impact to margins. Without the impact of PPA-related depreciation and amortization, ATM gross profit margin would be 30.2% and operating profit margin would be 20.1%.
On page 10, you'll find a graphical representation of our pro forma ATM P&L. On page 11 is our pro forma ATM revenue by market segment. The market segments were unchanged as compared with the previous quarter, and though the automotive segment is not separately displayed here, it continues to outpace the other market segments performances. On page 12, you will find our pro forma ATM revenue by service type. There was a small move in which our wire bond products grew slightly faster than our other product lines. This was primarily the result of seasonality of underlying products. On page 13, you can see the second quarter results of our EMS business. During the quarter, demand was stronger than anticipated, driven by stronger than expected demand for both our traditional EMS and SIP services. Overall, operating conditions started improving halfway through the quarter.
However, China's COVID mitigation strategy continues to have spotty impacts throughout our EMS business. Though the situation is still somewhat dynamic, our EMS factories are poised and ready for the third quarter seasonal uptick. During the second quarter, EMS revenues increased TWD 5 billion or 8% sequentially, and increased TWD 17 billion or 35% year-over-year. Revenues were somewhat ahead of where we expected, primarily as a result of higher than expected SIP and traditional EMS business. Overall profitability for our EMS business improved, with gross margin increasing 1.2 percentage points to 10% and reaching our 4% operating margin target. On the bottom half of the page, you will find a graphical representation of our EMS revenue by application. The reduction in the communications-related segment is primarily due to seasonality.
On page 14, you will find key line items from our balance sheet. At the end of the quarter, we had cash equivalents, and current financial assets of TWD 79 billion. Our total interest-bearing debt was TWD 218.3 billion. Total unused credit lines amounted to TWD 312.4 billion. Our EBITDA for the quarter was TWD 35.2 billion. Net debt to equity was 50%. On page 15, you will find our equipment capital expenditures. Machinery and equipment capital expenditures for the second quarter in US dollars totaled $515 million, of which $290 million were used in packaging operations, $161 million in test operations, $53 million in EMS operations, and $11 million in interconnect material operations and others. We continue to provide our EBITDA in US dollars here as a reference.
We believe that the company's EBITDA relative to our equipment CapEx serves as a key financial performance metric for the company. For the quarter, EBITDA was $1.2 billion. Joseph Tung will now present our outlook. Joseph?
Thank you, Ken. Before I give the specifics, let me reiterate that, despite the macro challenges, including the pandemic and related lockdowns, Ukraine war, supply chain disruptions, global inflation, escalating energy costs, and segment inventory corrections, we were still able to significantly outgrow our second quarter estimates and reach close to our anticipated peak revenue for the year. As a result, we will see a more linearized revenue outlook and continue to see a higher operating cost structure for the remainder of the year. Having said that, our guidance for the third quarter is as follows. For our ATM business, in US dollar terms, our ATM third quarter 2022 business levels should be slightly above second quarter 2022 levels. On a pro forma basis, our ATM third quarter 2022 gross margin should be similar with our fourth quarter 2021 gross margin.
As a reference, our fourth quarter 2021 gross margin was 28.5%. Now, for our EMS business, in US dollar terms, our EMS third quarter 2022 business sequential growth should be similar with same period last year. Again, as a reference, the third quarter last year's growth rate was up 25%. Our EMS third quarter 2022 operating margin should be similar to second quarter 2022 levels. Our second quarter 2022 operating margin was 4.0%. Now, with that, we'll open the floor for questions.
If you have any questions, please raise your hand. When you ask questions, please hold two questions at a time. Thank you. Our first question is from Mr. Randy Abrams. Randy.
Okay. Yes, thank you. A good result. I wanted to ask the first question just a little more on the guidance. For third quarter, it's, I see ATM high base. Second quarter did well. The factor for the small sequential growth is that how much is capacity limited, that you're tight on capacity? How much is it you're seeing any moderation or slowdown in certain applications?
Our utilization, our loading remains to be high. We do see, this year, the revenue from a quarter-to-quarter basis seems to be more linearized. Because in the second quarter, we have better than expected revenue stream coming in. Although there are some inventory corrections in different sectors, but there are also other sectors that remain very, very strong. Overall, I think it's not a capacity constraints kind of situation, but it's the overall market movement that is causing the year to be a more linearized revenue stream.
Okay. In the prepared remarks you talked earlier about we'll get back to return to the first quarter, where we traditionally have that seasonality. I think like high single, low teens decline. If you could give initial view between that, fourth quarter. Some years, I think in middle of the year, you've had enough confidence to say, I see ATM would grow in fourth quarter. I guess for this type of year, do you still see linear like a, like from the high base, a little bit of growth? Or, would we start some inventory correction in fourth quarter?
I think for the second half of the year, again, the revenue seems to be more linearized. In fourth quarter, we're expecting at this point a similar quarter to third quarter, although there is still some opportunity for further growth on a sequential basis.
Okay. One further or two follow-ups on growth. For the applications, I mean, we've had a lot of commentary about almost two cycles, but the consumer cycle, smartphone, TV, PC, that's the softness. I'm curious where you're talking about inventory correction, is it those areas that you're seeing it? Then the other commentary on demand outlook. It looks like EMS after a strong second quarter, is also quite strong. Is that function of any shift like earlier build or additional projects this year or other factors driving the good strength on EMS?
Well, let me comment on the sector inventory control, and I think there's enough conversation out there to talk about the consumer sector inventory control. I will not further elaborate. We're seeing some sector still very constrained. Then in the data center, in the networking, high performance computing and of course, automotives. We continue to see this kind of a sector fluctuation or rotation. In terms of the EMS trends, in addition to our normal SIP projects, I think this year we do have a good amount of automotive and the other different type of EMS projects. The answer is yes, we do have some new wins.
Our next question is from Mr. Gokul Hariharan.
Yeah. Hi. Can you hear me?
Yes.
Thank you. My first question, could we talk a little bit about what we are seeing for utilizations by the three big capacity categories, testing, wire bonding, and for Advanced Packaging. Are we starting to see any slack in any of these areas right now?
I think the overall packaging utilization remains the same as first quarter at about 80%-85%. The same pattern will be going into third quarter as well. For testing, the overall utilization still remains to be above 80%, and that will also go into third quarter as well. In terms of packaging, I think the overall loading is higher in the more Advanced Packaging. Whereas wire bonding, it's still growing and it's still quite full. It's comparing to the Advanced Packaging. I think Advanced Packaging seems to have a stronger momentum.
Got it. Thank you very much. Also wire bonding, I think the industry overall and ASE also saw some increase in prices, given the tight capacity last year.
Are you talking about the?
Yeah. Given some customers going. Yeah, exactly.
Well, we're seeing a stable pricing environment. In other words, if you're asking, is there any structural negotiation on the price down, the answer is no. We do see a stable pricing environment across all of the service category. I think we expect that between Q3 as well as Q4. I think this scenario might well last into 2023.
All right. Thanks. Thanks very much, Tien. Maybe one more question on that front. I think you had long-term loading agreements with a lot of customers signed last year as well as early this year. Are any customers starting to discuss any changes to these, being similar to what we are getting in the foundry side, where there are some LTAs being renegotiated?
I think every company signing LTA are different form. When we sign the LTA, typically it's with the NPI loading agreement as well as the pricing stability for specific lines. We don't sign the LTA for some lines, but we do sign LTA for like 100% LTA for some lines. Right now it's a mix. To answer your question, the short answer is no, we don't have any customer coming back to renegotiate LTA.
Next question is from Mr. Szeho Ng of China Renaissance. Szeho Ng? Szeho Ng?
Hi. Sorry. Yeah. I muted my line. Yeah. Actually, two questions from my side. First one, regarding the dividend policy, yeah, because the company pay a pretty high dividend this year. I just want to know the company's dividend policy or philosophy going forward.
I think in terms of the dividend amount is higher than previous years is because in terms of the payout ratio is pretty much the same. I anticipate that we will continue this payout percentage going forward. Although part of the profit that we generated through the sale of the sites, some of that earnings, we will maybe have some percentage of that profit will be given out this in the following years.
Okay. All right. Yeah. Roughly speaking, what percentage of our capacity right now is under LTA agreement?
About 70%.
All right. Okay. That's good. Yeah, last one. Yeah, on the CapEx front, I think we are sticking to the TWD 2 billion figure for this year, but how about the percentage breakdown among different divisions?
CapEx percentage breakdown.
Okay. I think overall for this year, we're looking at the allocation between assembly tests, material and EMS. I think in terms of packaging, the overall percentage will be about 54%, down from 65% last year. We're increasing our tests CapEx from last year's 25% to this year's 30%. Material, we're also increasing our CapEx, and the percentage will rise from 2% to 4% this year. For EMS, we're raising that from 9% to 12% this year.
We have a question from Mr. Bruce Lu of Goldman Sachs.
Hello, can you hear me?
Yes.
Okay. Thank you for taking my question. The congrats, great result. I'm very impressed in terms of your gross margin for both EMS and ATM. I know part of that is due to the currency, but still the gross margin improved a lot. However, you are guiding for slightly decline in the third quarter for the ATM gross margin, with the currency is still very weak and, you know, the capacity utilization rate remain at the same level. Can you give us a little bit more color on that?
That's more. There are two factors involved. One is, for some of the product mix with some of the higher material contents, shipments will be higher. The other main reason is really the rising utility costs.
I see.
Which will give us over 50 basis points increase in terms of cost.
I see. Can I assume that, you know, if the utilization rate remain at the current level, the 28% and above will be a new norm for ATM profitability moving forward?
I think Ken mentioned, you know, our structural improvement in terms of our margin. I think if you go back past 10 years, I think the gross profit margin range is from 20% to mid-20%. I think with the scale enlargement, with the technology improvement, also a lot of the efficiency improvement, I think structurally we can anticipate a move out of that range from mid-20s to 30%.
We can expect somehow 30% gross margin next year?
Well, that's the current goal. Also bear in mind that with the raised range, the new range is including the PPA, which has about a 1% impact on our overall margin.
Understand.
Yeah.
Understand.
If it's apples to apples, there's another 1% increase in our structural margin there.
Okay. My next question is regarding the Advanced Packaging business. I mean, hypothesis as to the, like, if you know, if TSMC is willing to outsource some of their Advanced Packaging business to ASE, do we have any threshold in terms of the profitability or ROE whether ASE would take that kind of business or not? Is that even in the company's growing strategy?
I mean, it certainly is a part of the growth strategy, but we will not comment on any detail on this one.
Just a little bit of color, Tien.
No comment.
Okay. I understand. I'll go back to the queue. Thank you.
Thank you. Our next question is from Mr. Rick Hsu of Daiwa Securities. Rick?
Yeah. Hi, guys. This is Rick here. I'll look, I guess.
Hello, Rick?
Yeah.
May you restate your question again?
Can you hear me?
Yes.
Okay. Right. I got only one question for you. Outlook for the year 2023. I think the tech cycle, you know, how much the cycle will drop, nobody knows. Some of the industry leaders like TSMC and UMC, they still expect a growth year for 2023, regardless of the cyclical correction. I would like to have your view on your revenue growth for next year. Can you give us a little bit color?
As I pointed out, it's really too early to talk about 2023. If you really wanna ask, and, I can offer you a most likely scenario. Right? I believe next year is going to be a challenging year with a lot of uncertainty and headwind. For the overall industry, it won't perform as well as the previous few years. However, in that scenario, you will have some companies, some sectors doing better than the others. Certainly, I believe the company you have just mentioned, as well as ASE, we're striving to perform in the upper threshold of the average. In other words, it is possible, even though the industry is flattish, slightly up or slightly down. Some company will continue to outperform the others.
I think next year will be a perfect scenario to see the differential of technology, customer traction, as well as the economy of scale. I believe that's what some of the companies we're referring to. In terms of specifically the upturn, downturn, whether recession is gonna hit us, and that really is very, very difficult. I don't think anyone knows. However, I do believe if there's a downturn, it would be a good opportunity for us to demonstrate that differential. Thank you.
All right. Thank you so much, Dr. Wu. That's very clear.
If I may add, I think, you know, our goal, our target is continue to do 2x of the logic semi growth.
Okay. This sounds pretty good. Just one quick follow-up about your guidance. I think when you state EMS Q3, I presume that EMS Q3, quote-unquote, growth will be around 25%, right?
Correct.
Okay, great. Thank you. Thank you so much, guys.
Thank you.
Now we have a question from Mr. Frank Lee of HSBC.
Great. Thank you guys. Just wanted to ask, I guess, looking at the bigger picture, and you've talked about, I think, a consistent message, some pockets of weakness, some other areas still holding up. But I guess if we look at this as a whole, the areas of weakness, such as smartphones and PCs, looks like they keep getting worse and worse over the last couple quarters. Yet your overall outlook has been good. You know, the overall business hasn't been too negatively impacted. As you go into next year, I know there's a lot of uncertainties, are there any specific end market applications that would be a concern? Because it seems like no matter how bad consumer gets, you know, it's so far it's not impacting you guys in the overall industry, especially some of the other foundries.
Are there any particular areas of concern that would be a bigger problem from a demand point of view that you can share?
You know, it's very sensitive to talk about which sector, because we will be automatically linked to the company and the customer that we're supporting. If I wanna talk about bigger picture in general terms, and then from ASE's perspective, you know, we are preparing for the following scenario. I call this one of the worst case scenario is people continue to order, assuming that either gaining share or nothing's gonna change, and the consumer most likely will go back to the traditional purchase pattern. Now, that will not affect the Q3 and Q4 loadings, because we have to prepare for Thanksgiving, Christmas, and maybe the Chinese New Year.
Once we stocked up, if the end market demand really shows strong sign of weakness, then there will be a major correction. Chances are in very late December, most likely will be in the January, February timeframe. This is why the general assumption right now of all of the supply chain guys that I talk to are assuming we will go back to the normal seasonality in Q1 of 2023. Okay? This is really what we point out as the worst case scenario, but I really would like to give all of you some operational highlight. Now, if you recall, all of the supply chain guys, including ASE, we have not had a break for almost two and a half years. You also understand that we're short of equipment, we're short of the manpower.
Everybody's calling everybody throughout the whole supply chain, everybody, just the same parts. For any operation under such an intensive stress-stretch, it's like running marathon three times, there will be fatigue. When I talk to all the supply chain, including my guys, everybody is kind of desperate for a break, for seasonality. Seasonality, you know, people can take the necessary attrition, equipment upgrade, software upgrade, maintenance, and many customers are pushing us to establish additional automation line. You understand that to do automation line, it takes a lot of resource, a lot of qualification. Then the COVID-19, after two years, now the customers start traveling. The NPI becomes very fascinating, so we have a lot of very interesting multi-die co-package, all kinds of application.
All of the NPI will occupy critical resources for qualification and redesign on material, process, equipment, software, everything. We have not had the time to do the necessary things to prepare for the next five years of growth. This is the general scenario I'm painting to you. I don't think anyone is of exception. Everybody's concerned about seasonality. Seasonality is this is like a four seasons. Like, this is like night and day. This is something we grew up with, and this is something health-wise, biological-wise, it makes sense business-wise. We really welcome seasonality. We should not really look at, you know, after two years of COVID, we should assume the abnormal to be normal. I really would like to, you know, start laying out, yes, we'll go back to seasonality.
We will try to well utilize the precious seasonality to do the right thing, like what we have always been trained to do. By doing this, we're looking at structure improvement, our baseline, our capability, our efficiency, better NPI, better engagement. Then, you know, industry is really poised for the long horizon. If you talk to anybody today, people will tell you, "Well, next five, 10 years is fantastic, but 2023, we don't know." Just take it. 2023, we don't know. Future five years, fantastic. Now, if that is the scenario, what would you do as an operator? I believe if we put things really in the big picture, in perspective, you will see that this is actually a necessary transition back to the normalcy. Whatever it is, for the long haul, it will be better for the industry. That's my honest opinion.
I really appreciate that. That's a great answer. Can I just ask a follow-up then, just on what you've painted as a worst case, you know, or I guess maybe the base case as to what you said. You know, things start to normalize by beginning of next year, or at least it's a two or three quarter slowdown, because that's been the past correction cycles we've seen in the industry. We've also seen kind of an unprecedented two-year upturn. Inventory levels, if you collectively look at semiconductors, is probably at a multi-year high, which we've never seen so high before. Do you think it is normal to assume that it'll be just a normal two, three quarter?
I know it's a very difficult question, but given where we are coming off of such a high base, could it be a longer period or is it, are we too optimistic to think that it's, you know, a normal two, three quarters in light of the fact that everything is at extremely elevated levels?
You know, the Joseph Tung and Ken Hsiang are signaling, asking me not to say it.
Okay.
I think I want to say it. You know, if you look at how bad the downturn is gonna be, please consider the following fact. I think if you look at the PC number, you know, whether commercial or personal, consumer or commercial, the baseline is moving up. That's a fact. If you look at the wireless, I really do not wanna talk about any customer. What we have seen is the wireless inventory control started in January of this year. I've never seen people jump in so early. You know, yes, you know, we have our agility, we're shrewd. Our customers are more shrewd.
I mean, you don't know, but like, you know, we used to be like 30%, 40% overbooked, sometimes 50% overbooked. In the last few quarters, things have changed dramatically starting January. The inventory control has been going on for two quarters already. The question now is the, I mean, Q3 additional inventory control by some sectors, Q4, another control. You believe this will last all the way to the second half of next year? Again, nobody knows. I only want to offer perspective. I believe the baseline, because of the COVID, people travel differently, people use Zoom differently, people use PC differently. The baseline has absolutely moved up, I'm convinced. The question now is, with all of the baseline moving up, as with inventory control, as with everything that's going on.
Right now we just don't know what it is. If you really ask my honest opinion, I think next year, you know, will be a mild adjustment. I mean, if you look at even the U.S. Fed. I mean, yesterday the market moved up, and then they raised three basis points. You hear everything. Again, you know, I don't have a crystal ball, and I just wanna share with you what our internal conversation has been. People are sharing extreme thoughts that next year is gonna be a holocaust. Terrible. On the other hand, look at the signs. Look at all of the NPIs and look at all of the booking. Look at all of the substrate. Look at all of the automotive guys, what a long waiting list.
I mean, I talked to a lot of customers, the way they're aggressively booking capacity, it really makes you think, "Why?" Again, by next January, we will see better. Then we can compare to my notes six months ahead of time. We'll see. Okay, Joseph asked me absolutely not to say anymore. I stop.
Now we have a question from Mr. Bruce Lu of Goldman Sachs.
Hey, I wanna ask one more question about the, you know, multi-year growth, which is 2x of, semiconductor growth. Can you break it down in terms of like what is the dollar content growth, what is the shipment growth, and what is the share gain in a way that we can better understand it? Or can you provide something like, you know, TSMC has been saying that their dollar content will increase by mid- to high-single-digit in the next few years. Can we have that kind of quantitative guidance for your growth for the next few years?
My apology. We don't have this kind of detailed dollar breakdown in terms of the end product of the customers. Because our customers are too diversified, and it's hard for us to track. In terms of, I mean, organic growth is easy. We also have the technology, which means that for the same unit, we have to start adding more value because of the complexity. On top of that, the share gain actually is not difficult to comprehend. If you look at the supply chain for the last two years, and you really ask question, where were the bottlenecks? You will understand that OSAT is hardly the bottlenecks. We have other bottlenecks.
Therefore, to have a more secure supply chain, a lot of the end customer, including automotives, are demanding to have second source. The second source being moving to the foundry and the OSAT. In terms of market share gain, we know we have market share gain, and that is giving us more confidence in terms of how do we have a stability moving forward, even though you will have a sector rotation up and down.
Okay. Can you say that the market share, again, is more important than your additional value add to your product, i.e. the dollar content growth? Which one is the more important factors?
I think the safer answer is they're all important because I really cannot give you. I'm sorry, you know. Yeah, I don't want you to-
Okay.
... misuse the information. I don't wanna mislead you.
No, I think it's very, very critical that we have a very, very unique scalability. I mean, when the time is uncertain, I think all the customers will feel for security and also fight for quality. I think that's what we have. I mean, the whole industry will continue to see unit growth, will continue to see IC content growth. We will continue to see increasing outsourcing. Even like automotive in the past, very little percentage of the auto parts are being outsourced. Now because of the customers' request, more outsourcing in the automotive is happening as well. We're seeing a lot of growth in the IoT.
We're seeing HPC, a lot of different categories that are still going strong. I think the overall unit continues to lay a very strong support for our business. Given our leading position in the industry, you know, market share gain is a very natural thing for us.
Just to put some additional color to the last two years, because now as the head of operations, you have to understand that in the last two years there's a lot of like commodity type of part number looking for volume, which is very critical. In the last two years, there are very difficult parts, like in automotive, that not everyone can do it. Then ASE was constrained by equipment delivery, and we're also constrained by manpower because of COVID. If you look at the number of parts that we shipped, and also if you really look into it, the content, there has been. We made a lot of friends, let's put it this way.
In an extremely difficult condition, we were able to, you know, just pull it together and then really start shipping in record time a highly complex high-volume parts using the fully automated line. That gained us a lot of friendships, which is rewarded by additional NPIs. When we look at the whole scenario, yeah, we do understand the installed capacity, we do understand our competitor. You know, with everything considered, and I think, you know, especially in 2023, I think it'll be interesting just to see, you know, all of the friendship, how much that is worth.
Can I ask you in a different way? For the last two years, obviously you are growing, you know, 2x than semi growth. For the last two years, at least you have some computational number that how much is coming from the share gain, how much it was coming from the, you know, higher value add you provide to your customers.
Yes.
At least for the last two years.
Well, when I don't know the number, I just tell you 50/50.
Okay. Understand that. Thank you.
We have a question from Mr. Randy Abrams of Credit Suisse. Randy?
Okay. Yeah, thanks for the follow-up. I wanted to ask on the capacity expansion, if you could clarify the announcements you've made in SPIL in terms of like timing and how much you're bringing on. Second part to that, we've had all these, CHIPS Act, Europe, now U.S. may go ahead, India. Is anything shifting in thinking? Because it looks like you have a lot on deck for expansion in Taiwan.
All right. Well, I'm not sure you're talking about the U.S., the upper house just passed the CHIPS Act. Is that what you're talking about? FABS? Is that more?
Yeah. It's not through all the way, but yeah, the upper, the Senate passed.
Right.
There's quite a bit of money in there, including some could go to packaging.
Right. I mean, obviously our customers and our governments, plural terms, are all pushing us to do the, you know, some investment. Our answer is the following. We would like to make that investment in different geography just to satisfy the supply chain security and everything. We need to understand precisely what are the requirements. Until today, we have not been able to narrow down to the precise investment that we need to make in order to make the supply chain security better. Now, the support, the funding is always good to have. However, we will not make any investment because of the funding or the subsidies.
We will make investment if we understand precisely the requirement in detail such that we can perform to that expectation. The short answer is in the short term, no. We're not engaging in any kind of capacity investment in any geography outside of our current jurisdiction. In terms of the SPIL investment, I think SPIL is building additional buildings in anticipation of the next few years of our ramp up. Then I think ASE is also acquiring land and building facilities. Because the facilities and the buildings are on a different timeline. They're on a different timescale compared to the business, the ups and downs.
We also wanna make sure the infrastructure is ready, and then we would deal with the business, the equipment, and the qualification accordingly.
Okay. Makes sense. If you could talk to this. You've mentioned a few times the multi-die, the co-packaged. In the past, there was a lot of focus, which was a fair amount of EMS. Could you go a little more into the application of the products, driving this and how big it is for the overall, if it's more IC ATM, how big it is?
Okay. When we talk about the multi-die co-package, my apology for all of these terms, because it is different. Because we always talk about the multi-chip module, the MCM. The MCM tends to be multiple chip stay using the similar method to be attached to one module. When I talk about the multi-die co-package, that means it could be multiple package, multiple die on multiple package, and these two put it together. When you think about this, then your question is relevant. You know, this kind of multi-die co-package, should that be going to the ATM business or should that go to the USI business? The question really is, the answer is yes, both. It really depends on the density, the complexity, and also the feature size. It also depends on the logistics.
The ATM can do it. The very fine pitch, a complicated ATM will do it. The little, whatever the EMS can handle, the EMS will handle. In terms of application, we see this in all applications, especially when you talk about automotive. The automotive used to be. I mean, if you look at the Tesla, for example, right? Their design always to be compared to the old automotive design, the old automotive design was like one microcontroller will control the radio and then the CD and then the windows. One chip will do, you know, one.
One thing.
One chip does one thing. If you look at how Tesla designs things, one chip does 20 things. In order to do this, you really have to start thinking about the concept of multi-die package and the managing by the firmware, software, and everything around it. In the future, like, you know, not only all of the combustion cars are going this way, the electric vehicles are absolutely going this way. That's just one example. When you think about multi-die package, I think, you know, you bundle the power management chip with ASIC, controller, memory, they're not necessarily the same. I mean, wireless chip, then you have linear, analog, you have digital. They're not just die. They can be die.
They can be dies on a package, and then they somehow bundle all of the packages together. Things become a little bit more creative now. I think it covers all of the applications, and this is something that we believe will offer us more value and room for imagination in the future.
Okay, great. Thanks a lot, Tien.
There is no more question.
Thank you, guys. Yeah.
Okay. If there's no more question, we will end the call today. Thank you very much.
Thank you very much.
We'll see you next quarter.