AAcquired
← All episodes
20 November 2017

Qualcomm - Broadcom

2Frameworks
12Insights

Listen

Frameworks in this episode

Insights & moments

The myth-busts, hot takes, explainers, and tools worth keeping.

Hot Take· 2

Hot Take26:00

Why Broadcom Bid While Qualcomm Was Under Legal Pressure

The hosts interpret Broadcom's offer as an attempt to purchase a valuable company while litigation and regulatory disputes depressed its share price. Qualcomm had recently traded near the proposed price, but Broadcom could present that price as a premium after the decline.

  • The FTC was suing Qualcomm
  • Apple and Qualcomm were litigating
  • External pressure had depressed Qualcomm's shares
  • The hosts expected Qualcomm to demand a higher offer

“boy, there's an opportunity to get a great company for pretty cheap because they're going through this sort of rocky time.”

Ben Gilbert · 26:30

“Broadcom being opportunistic from a time frame perspective right now.”

Ben Gilbert · 27:00
#acquisition#valuation#litigation#qualcomm
Hot Take47:00

Could Regulators Accept One Supplier Controlling 65% of Handset Components?

The acquisition's industrial logic also created its largest regulatory problem. The hosts cite an estimate that the combined company would supply 65% of handset semiconductor bill-of-materials value and question how Broadcom could persuade US and EU regulators that the deal was not anticompetitive.

  • The combined supplier could control 65% of handset semiconductor content
  • US and EU regulators apply different competition standards
  • Patent royalties add another concentration concern
  • Regulatory approval remained a major assumption in the hosts' grade

“if they combine, then 65% of the bill of materials um of handset semiconductors will come from one single manufacturer.”

Ben Gilbert · 47:30

“I will be so curious what the story is to regulators of why this is not anti-competitive to push it through.”

Ben Gilbert · 50:30
#antitrust#regulation#market concentration#semiconductors

Explainer· 5

Explainer01:00

How Fabless Chip Companies Build Hardware Without Owning Fabs

Qualcomm and Broadcom design semiconductor components but contract out their physical production. The model emerged because advanced fabrication facilities became so expensive that only a small group of companies could operate them efficiently.

  • Fabless companies specialize in chip design
  • Foundries perform the physical manufacturing
  • Advanced fabs can cost more than $10 billion
  • TSMC, Samsung, and Intel are major manufacturers

“both effectively fabless semiconductor companies.”

Ben Gilbert · 01:00

“it costs 10 billion plus dollars to create one of these fabs.”

Ben Gilbert · 09:30
#semiconductors#fabless#foundries#manufacturing
Explainer06:30

Why Machine Learning Revived Semiconductor Startups

GPU demand from machine learning transformed NVIDIA and encouraged investors to fund specialized chip startups again. Foundry access lowered the manufacturing barrier, although advanced design still required unusually large engineering teams and financing rounds.

  • NVIDIA rose more than tenfold in roughly two years
  • GPUs suit the matrix operations used in machine learning
  • Intel acquired machine-learning chip startup Nirvana
  • GraphCore raised $110 million in just over a year

“It turns out, you know, graphics cards are good for more than just graphics.”

David Rosenthal · 07:30

“this company's raised $110 million dollars in just over a year”

David Rosenthal · 08:30
#machine learning#nvidia#startups#venture capital
Explainer12:30

What Qualcomm and Broadcom Actually Supply Inside Smartphones

Teardowns of the iPhone X and Pixel 2 XL reveal that even vertically integrated phone makers rely on numerous external semiconductor components. Qualcomm supplies modems, transceivers, processors, and power circuitry, while Broadcom supplies components such as wireless-charging and touchscreen controllers.

  • Apple designs its A-series processor internally
  • Apple dual-sourced LTE modems from Intel and Qualcomm
  • Broadcom supplied several iPhone X components
  • The Pixel 2 XL contained numerous Qualcomm components

“There's actually plenty of Qualcomm and Broadcom components in uh in the iPhone.”

Ben Gilbert · 13:00

“everybody is a friend of me everybody's a competitor everybody is an in co-opetition.”

Ben Gilbert · 14:30
#iphone#pixel#qualcomm#broadcom#components
Explainer27:00

The Debt, Side Deals, and Jurisdiction Changes Complicating the Bid

The proposed acquisition sat amid multiple pending semiconductor deals and relied heavily on cash financing. Broadcom was buying Brocade, Qualcomm was buying NXP, Silver Lake offered convertible debt, and Broadcom planned to move its legal address from Singapore to Delaware.

  • Silver Lake committed $5 billion in convertible debt
  • Broadcom's offer mixed cash and stock
  • Broadcom was acquiring Brocade
  • Qualcomm was acquiring NXP
  • Broadcom announced a legal relocation to Delaware

“They committed $5 billion in convertible debt to finance this acquisition for Broadcom.”

Ben Gilbert · 27:00

“Broadcom announced that it will relocate its legal address from Singapore to Delaware.”

Ben Gilbert · 29:30
#deal structure#debt#regulation#m-and-a
Explainer37:00

How Apple's Vertical Integration Threatens Horizontal Chip Suppliers

Successful device makers increasingly design important components in-house, reducing the addressable market for independent semiconductor suppliers. This creates a graduation risk: the better a supplier enables a customer to become, the more capable that customer may become of replacing it.

  • Apple established a precedent for internal chip design
  • Horizontal suppliers lose content when customers integrate vertically
  • The risk rises as customers become larger and more capable
  • Embedded and IoT customers may be less likely to design chips

“when you become hugely successful, you can graduate away from the chipmakers into doing it in-house.”

Ben Gilbert · 37:00

“if you enable your customers to be too successful, will they start doing this in-house?”

Ben Gilbert · 37:30
#vertical integration#apple#supplier risk#smartphones

Story· 2

Story17:00

Why Today's Broadcom Is Really Avago and Part of HP's Legacy

The company operating as Broadcom resulted from a merger with Avago, after which Avago adopted the better-known Broadcom name. Avago itself originated as Hewlett-Packard's semiconductor division and passed through multiple divestitures and combinations.

  • Avago merged with Broadcom
  • The combined company retained the Broadcom name
  • Avago began as HP's semiconductor division
  • Its lineage reaches back to early Silicon Valley

“the company today known as Broadcom really is a Vago. They took the Broadcom name.”

David Rosenthal · 17:30

“Avago uh started life itself way back in the day as HP's semiconductor division.”

David Rosenthal · 17:30
#broadcom#avago#hewlett-packard#history
Story22:00

How Qualcomm Left Handsets and Built a Chip-and-Licensing Business

Qualcomm progressed from cellular base stations to branded phones before selling its handset operation to Kyocera. It then concentrated on chipsets such as Snapdragon while continuing to earn billions from licensing foundational cellular patents.

  • Qualcomm once manufactured branded mobile phones
  • It sold its handset business to Kyocera
  • The company shifted toward handset chipsets
  • Licensing generated about $8 billion in the cited year

“Qualcomm's 10k last quarter or last year was 15.4 billion in revenue from equipment and services and eight billion from licensing.”

Ben Gilbert · 22:00

“They sold the the handset business to Keosera.”

David Rosenthal · 23:00
#qualcomm#handsets#snapdragon#licensing

Q&A· 1

Q&A39:30

Can Capital-Intensive Semiconductor Startups Fit Venture Capital?

The hosts examine the tension between Sequoia's warning about capital intensity and its investments in semiconductor companies requiring enormous early financing. The potential winners can become exceptionally valuable, but each investment consumes enough capital to undermine normal portfolio diversification.

  • Capital-intensive markets can produce huge winners
  • Most heavily funded competitors may still fail
  • Large checks reduce the number of portfolio bets available
  • Bigger venture funds make such investments easier to absorb

“capital intensity usually produces nightmares.”

David Rosenthal · 40:00

“you just can't take that many, you can't put that many of those companies in a portfolio, right?”

David Rosenthal · 43:00
#venture capital#capital intensity#portfolio strategy#semiconductors

Takeaway· 2

Takeaway09:00

The Foundry Infrastructure That Made New Chip Designers Possible

Consolidating fabrication into specialist foundries gave semiconductor startups a manufacturing platform they could build upon. The hosts compare this enabling infrastructure to the electric grid, UPS, telephone networks, and the internet that supported Amazon's creation.

  • Specialist foundries absorb enormous capital costs
  • Design firms can innovate without building factories
  • Shared infrastructure creates new entrepreneurial possibilities
  • Semiconductor startups remain more capital intensive than software firms

“So suddenly you have this platform where you don't need the actual capital to to do the manufacturing yourself.”

Ben Gilbert · 09:30

“everyone that came before to basically build platforms that you could do innovation on top of.”

Ben Gilbert · 10:00
#infrastructure#innovation#foundries#platforms
Takeaway48:00

Why Acquired Gave the Proposed Qualcomm Acquisition a B

The hosts view the deal as strategically sensible but not transformative. Broadcom could gain scale and acquire Qualcomm during a period of depressed valuation, yet the transaction offered fewer exciting product or technology benefits than landmark acquisitions such as Instagram.

  • The hosts assumed a somewhat higher final purchase price
  • Broadcom could realize meaningful economies of scale
  • The combination appeared necessary rather than transformative
  • Both hosts settled on a B grade

“I'm gonna go, I don't know what to do other than a B, right? Like it's not transformative.”

David Rosenthal · 49:00

“So I'm I'm with you on a B, I'll agree on that.”

Ben Gilbert · 50:00
#acquisition grade#broadcom#qualcomm#valuation