Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Foundries are taking on more manufacturing for integrated device manufacturers (IDMs), but that does not mean IDMs are disappearing or one foundry will control the whole industry. The shift reflects the cost and scale of running advanced fabs: many chip companies now outsource some production while keeping selected manufacturing in-house. TSMC’s reported share of its expanded “Foundry 2.0” market rose from 28% in 2023 to 34% in 2024, while Analog Devices said third-party foundries supplied more than half of its wafer requirements in fiscal 2024. Those figures show a meaningful shift, not a universal takeover.
What is changing between IDMs and foundries?
An IDM designs chips and manufactures at least some of them in its own fabs. A foundry manufactures chips for customers that may not own fabs, as well as for IDMs that choose to outsource some production. In practice, the boundary is increasingly flexible: an IDM can retain its own processes and facilities while buying outside capacity for particular products or technologies.
GlobalFoundries, citing IC Insights, reported in a 2021 SEC-filed presentation that more than 33% of semiconductor manufacturing was outsourced to foundries in 2020, compared with approximately 9% in 2000. The presentation attributed the change to rising manufacturing costs and the difficulty of earning a return on the capital invested in fabs. Its conclusion was that it had become increasingly difficult for IDMs to scale manufacturing profitably on their own.
The hybrid model in practice
Analog Devices illustrates the approach. In its fiscal 2024 filing, the company said third-party foundries, including TSMC, supplied more than half of its annual wafer requirements; it produced the remainder internally. That is not a full exit from manufacturing. It is a way to combine internal production with outside capacity where a foundry’s scale or technology makes sense.
#1 Best Overall
Why are chip companies outsourcing manufacturing?
Fab economics favor scale and utilization
Fabs require large, continuing investments in equipment, process development and operations. A manufacturer serving many customers can spread those costs across more production and has a broader opportunity to keep capacity utilized than a company relying only on its own chip demand. GlobalFoundries argued in its 2021 SEC-filed presentation that foundries have the volume needed to generate returns on the capital required for manufacturing.
Leading-edge processes demand sustained investment
Developing advanced process technology requires ongoing research and investment. TSMC reported that research and development spending was 7.1% of revenue in 2024. In its 2025 report, it said processes at 7nm and more advanced accounted for 74% of wafer revenue in 2025, up from 69% in 2024, and that 2nm entered high-volume manufacturing in the fourth quarter of 2025. These are TSMC’s reported figures; they describe its own business, not the industry as a whole.
AI demand and advanced packaging add to the appeal
TSMC links AI deployment with strong demand for advanced-node chips. Large AI systems also need ways to connect processing, memory and other components efficiently. TSMC says it is developing advanced 3D packaging technologies including CoWoS, InFO and SoIC to support large-scale interconnectivity. The combined demand for advanced chips and packaging can favor manufacturers that can invest in both process technology and the supporting production ecosystem.
A wider customer base can improve the foundry model
Foundries can serve fabless chip designers, large technology companies and IDMs. That broader pool of customers can support higher production volumes and spread fixed costs across more products than a single IDM might achieve on its own. It also makes foundry capacity strategically important to companies that do not want to build or operate every manufacturing capability themselves.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
What do the market-share figures actually show?
The figures below describe different things and should not be treated as one continuous market-share series. TSMC’s “Foundry 2.0” measure is broader than the traditional foundry market, while the outsourcing figures track manufacturing arrangements over time.
| Figure | What it measures | Source and qualification |
|---|---|---|
| 34% in 2024, up from 28% in 2023 | TSMC’s share of revenue under its expanded Foundry 2.0 definition, which includes logic wafers, packaging, testing, mask-making and non-memory IDM activity. | TSMC, 2024 reporting. This is not a share of wafer manufacturing alone. |
| Close to US$250 billion in 2023, compared with US$150 billion under the narrower traditional foundry definition | Estimated size of the expanded Foundry 2.0 market versus the traditional foundry market. | TSMC’s 2023 estimate. The two market sizes use different definitions. |
| More than 33% in 2020, compared with approximately 9% in 2000 | Share of semiconductor manufacturing outsourced to foundries. | GlobalFoundries’ 2021 SEC-filed presentation, citing IC Insights. These are historical figures, not current estimates. |
| More than half of annual wafer requirements in fiscal 2024 | Wafer supply obtained from third-party foundries, including TSMC. | Analog Devices’ fiscal 2024 filing; the company said it produced the remainder internally. |
| 74% in 2025, up from 69% in 2024 | Share of TSMC wafer revenue from processes at 7nm and more advanced. | TSMC’s 2025 report. This measures its revenue mix, not its share of all advanced-process production. |
The safest conclusion is that outsourcing has grown and foundries have gained commercial importance. The evidence does not show that every IDM is outsourcing, that every market segment is moving at the same pace, or that TSMC owns the share represented by the entire Foundry 2.0 category.
What does “Foundry 2.0” include?
“Foundry 2.0” is TSMC’s expanded market definition. It includes traditional wafer foundry work alongside packaging, testing, mask-making and manufacturing activity at non-memory IDMs. That wider scope is why TSMC estimated the 2023 market at close to US$250 billion, rather than US$150 billion under the narrower traditional foundry definition.
The distinction matters when interpreting TSMC’s reported 34% share for 2024. It is a share of the expanded category, not a claim that TSMC made 34% of all semiconductor devices or 34% of all wafers. The category also includes other manufacturers and service providers, so the statistic does not establish a monopoly.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Are foundries replacing IDMs, and does TSMC have a monopoly?
Neither conclusion follows from the evidence. Outsourcing can grow while IDMs continue to design chips, run selected fabs and keep specialized processes in-house. Analog Devices’ fiscal 2024 mix—more than half of wafer needs supplied externally, with the rest made internally—is a concrete example of coexistence.
TSMC is a major foundry, particularly in advanced manufacturing, but the broader Foundry 2.0 category includes Samsung, Intel and other IDM manufacturers as well as packaging and testing providers. Customers can also have reasons to qualify more than one supplier, including access to capacity, supply security and regional requirements. A large reported share in one defined market does not establish control over every process node, specialty product or region.
What could limit further foundry gains?
Capacity dependence can reduce customer control
Using an outside foundry can expose a chip company to its supplier’s capacity decisions, schedules, yields and costs. Analog Devices warns that foundries also serve competing customers, so a demand spike can tighten capacity. Outsourcing may reduce the need to fund a fab, but it does not guarantee priority access to production.
Geographic concentration creates supply risk
Analog Devices warns that tensions across the Taiwan Strait could disrupt TSMC operations. Separately, GlobalFoundries reported that about 77% of foundry revenue in 2020 came from wafers manufactured in Taiwan or China, highlighting the geographic concentration of supply at that time. That historical statistic is not a current geographic breakdown, but it illustrates why location and resilience matter alongside cost and technology.
Best Value
Demand varies by process and industry
AI is supporting demand for leading-edge chips, but demand is not uniform across the semiconductor market. TSMC has also described weaker or correcting demand in some automotive, industrial and IoT mature-node markets. A foundry’s strength in advanced logic does not mean it dominates every mature-node or specialty process segment.
Regional diversification is expensive and takes time
TSMC’s 2025 report describes manufacturing projects outside Taiwan: its first Arizona fab entered high-volume production in the fourth quarter of 2024; its second Arizona fab was expected to enter high-volume manufacturing in the second half of 2027; and construction began on a third Arizona fab in 2025. The report also says the first Kumamoto fab began volume production at the end of 2024, a second Kumamoto fab is under construction, and a Dresden specialty fab is progressing. These projects can broaden geographic options, but building and operating capacity in multiple regions is costly and complex; expansion does not by itself eliminate concentration risk.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How should companies assess foundry dependence?
Market share alone is not enough to judge whether a foundry or an IDM is better positioned. For a company choosing where and how to manufacture, useful comparison points include:
- Process leadership and yield: whether the supplier can make the required chips at the needed performance, yield and scale.
- Capacity and utilization: how much production is available, how reliably it can be scheduled, and whether capacity is stretched during demand spikes.
- Packaging integration: whether the manufacturing partner can support the packaging and interconnect needs of the finished system.
- Geographic resilience and support: where production is located, what disruptions could affect it, and what regional investment or government support may matter.
- Customer concentration and trust: whether the supplier’s customer mix creates competing claims on capacity, and whether the buyer is comfortable relying on that relationship.
- End-market and node exposure: whether a company depends on AI-led advanced-node growth, mature-node demand, or a mix that behaves differently over the cycle.
What is the likely direction of the industry?
The evidence points to continued foundry growth alongside a hybrid IDM model, rather than the disappearance of in-house manufacturing. Foundries can gain work when their scale, advanced processes or packaging capabilities are difficult for one chip company to justify alone. IDMs can still retain internal production where proprietary processes, specialized requirements or control over supply make it worthwhile. How far the shift goes will depend not only on technology and demand, but also on available capacity, regional resilience and the economics of each product segment.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
