Hacker News new | comments | show | ask | jobs | submit login
Qualcomm rejects Broadcom's $103B takeover proposal (reuters.com)
349 points by rbanffy 7 months ago | hide | past | web | favorite | 90 comments

> the Board has concluded that Broadcom’s proposal dramatically undervalues Qualcomm and comes with significant regulatory uncertainty

Qualcomm's stock took a dive since Apple (25% of their revenue) said they might not need them for their next models. They're probably gonna wait to bounce back.

Isn't there a big lawsuit between Qualcomm and Apple pending?

There are many lawsuits and trade dispute filings pending from both sides. Qualcomm is also under anti-competitive practice investigations in a number of places.

It's going to take years to sort out and will cost a lot - stock has already tanked 2x what Apple's lost revenue was worth, so the market thinks this is not only more damaging than currently told, but also that they won't recover

I can't see them surviving years without being taken over

Well, the board rejects it, but the shareholders can still sell, no? Hostile takeovers have worked in the past, e.g. Oracle's acquisition of PeopleSoft.

There was a brief dip in share price earlier today, but it seems to be recovering already.

Broadcom's offer was a pretty generous one, and if they were to make it to minority shareholders, they'd probably be successful. Unsure how much of qualcomm's capital is in the hands of the board or their associates..

No one opens with their best offer though.


Sounds pretty damn good to me.

I'd rather have one-hundred-and-twelve-thousand-million-dollars. Deal?

Best I can do is 3.50

(Sorry HN “this is not reddit” comments ensue)

(Get a sensor of humor you drones)

I cant see it happening. Most of the large share holders (largest is 2%) are index funds and EFT's.

I don't really know, but it isn't impossible to do a takeover when index funds own a lot. If ISS is on the side of the acquirer, the index funds are likely to go with what they say.

I'm actually a bit surprised that they don't just abstain, so that their vote doesn't count. I mean, if you're going to be a passive investor, actually be passive. Voting in favor of the management or ISS all the time is not passive.

I'm just confused at what the International Space Station has to do with it? (Yes, I DDG'd (duck duck go'ed) ISS, guessing it is some trade association or society, but I came up with International Space Station, and a very long list of other possibilities).

Wikipedia's disambiguation page is typically want you want in scenarios such as this:


Near the bottom of the list, you'll find "Institutional Shareholder Services, a proxy advisory firm":


It would have been probably been rejected by anti-trust regulators.

It's far from over!

Broadcom/AVGO likely assumes that their bid to takeover Qualcomm would not be met with smiles. Broadcom can convince Qualcomm shareholders to replace the board with one that will accept the offer (or subsequent offers). Once the friendly board is in place, they might halt the NXP acquisition (forfeiting some good-faith deposit to NXP) and convince regulators that they would divest Qualcomm's IP from their semiconductors. This would make regulators happy and it would set the stage for them to court AAPL for future design wins.

> Broadcom can convince Qualcomm shareholders to replace the board with one that will accept the offer (or subsequent offers).

Is that even legal?

Totally. That's the definition of a hostile takeover. In principle, the board just represents the shareholders, and if they're not doing what the shareholders want then they have the absolute right to replace them.

In the US? That would be very surprising.

Anti-Trust law requires consumer damage under the modern interpretation and its hard to see how a globally competing business would satisfy that.


I feel this might have something to do with the anti-trust case.

Very possibly - and Qualcomm seemed to agree. They're board mentioned "regulatory uncertainty" in their reasoning.

what makes you think so? I mean it should be, but I don't think the govt is currently being run by a sane administration.

Both of those companies are global and would need to pass antitrust in the EU, China, and Japan as well. I find it highly unlikely Europe would agree to it.

Nothing to do with the administration. How the law is applied deepens on judges, we are not in the progressive area (they were far, far more racist then Trump).

Are Qualcomm and Broadcom subjected to EU antitrust laws when it comes to acquisitions? Or just limited to when one side is registered in a EU country?

I think that any acquisition between companies that have a relevant market in the EU may be subject to scrutiny.

Effectively both are in all jurisdictions because they sell their parts globally. So, the EU certainly would be able to reject the acquisition.

What does that mean if the EU rejects it? They are American companies, no? If the US doesn't reject it, why can't they merge? Or can they, but the EU rejection means they can't sell things in the EU or other ramifications?

It's curious to me because China does multiple things that wouldn't be allowed in the EU. Like working far too closely with the government can be allowed for Chinese companies, but mergers between them can still be rejected by the EU?

The landmark case is the GE-Honeywell merger, which the EU blocked after the US DoJ had already approved. In it, the Commission said they had the authority it involved at least two "companies with a combined turnover worldwide in excess of €5,0 billion and European sales of at least €250 million".

I'm not sure what would be the procedure if they went ahead anyway, but when you control €500M+ in sales in your jurisdiction, you have a strong position.

Great article for anyone interested.[1]Thanks for making me aware of this.


Since QCOM hasn't given back the 13% pop the offer provided, it seems shareholders are expecting BRCM to come back with another, higher offer.

AVGO and yes most expected Qualcomm to reject the offer to solicit superior ones. This isn't news to anyone, but rather a formality.

Bad move by Qualcomm. Now is the best time to cash out. Apple, Samsung and LG have been working on their own chips, and will have no need to go through Qualcomm in the next 1-2 years. Seems like ego is getting in the way of the solid decision making

Qualcomm just released a server chip, broaching a new market for them. And that chip is a serious pain in the ass for Intel. Why would they want to be bought by Avago, which itself divested of Broadcom’s Arm server effort earlier (which recently launched as well anyway!)

I think they are probably going to mess up with the new chips like they did with their OS and Qualcomm will be back up in sometime.

they just want a better offer

How many competitive LTE chipset manufacturers exist?

It's worth noting that Broadcom does not make one. Qualcomm and Broadcom overlap in WiFi + Bluetooth.

I know of three LTE modem manufacturers: Qualcomm, Intel, Mediatek.

Doesn't Huawei/HiSilicon make one? (apparently designed by GCT Semiconductor)

Yes, I think you're right.

Doesn't Samsung make its own LTE chips, too, for Exynos?

I also wish Samsung would sell its Exynos chips to other smartphone OEMs. They've hinted in the past that it's Qualcomm the one blocking them from doing that with patents. Regulators should take a closer look at that.

Meizu uses Exynos chips on their smartphones, looks like they are the only OEM so far.. http://www.samsung.com/semiconductor/minisite/Exynos/Newsroo...

That's what patents are for. If Samsung is too cheap to license them then they don't get.

Samsung is the largest customer and licensee of Qualcomm chips and patents.

The problem here isn't that Samsung doesn't want to license Qualcomm's patents, but that Qualcomm DOES NOT license their patents to their baseband competitors (eg, Samsung, Mediatek, Intel); instead Qualcomm wants to collect royalties from the end-device makers (eg, HTC, Samsung Mobile).

This is precisely why anti-trust/competition regulators have been going after Qualcomm recently.

Patent holders don't have to license anything if they don't want to. That's how patents work.

It's not like QCom are NPE trolls. They actually did all the ground work of creating CDMA as a viable technology. The patent system is there to reward them for their research efforts.

Qualcomm's wireless patents are Standard Essential Patents (SEP) and FRAND (Fair Reasonable Add Non-Discriminatory) encumbered.

In other words, Qualcomm voluntarily declared their wireless patents to Standard Setting Orgnizations (SSO) such as ETSI, ITU, etc to become part of standards that all chipmakers must implement. Qualcomm is, under their agreement with SSO's, required to license their wireless to all willing licensees at FRAND terms. In exchange for giving up their monopoly, Qualcomm gets guaranteed income from all uses of wireless standards such as LTE in baseband chips, or mobile phones.

MediaTek is only suitable for mid-tier phones (at least when I was in the OEM industry). Exynos is only for Samsung. A lot of OEMs try to make their own LTE chips, but it's not easy.

Qualcomm is still very much relevant in the phone industry.

Meizu has been using Samsung's Exynos chips for years now.

I think there are plenty of radio modem makers/competitors now. It's not to say that it's easy to make your own chips, but the real problem here is that Qualcomm does not believe in licensing their wireless patents to baseband makers. Qualcomm wants to collect royalties from OEM's customers instead for greater licensing fee.

Some might argue that at this point, it’s their only real relevance, yet a hugely valuable (and legally protectable) relevance. And they know it, if their recent advertising messaging is any indication.

Does anyone know who would be making the chipsets that come after LTE? Do patents for each round of network technologies change which companies make them? Or does it stay the same?

Really, I'm wondering what happens whenever 5G or whatever comes after 4G/LTE comes around. And in how long approximately that will be.

> Does anyone know who would be making the chipsets that come after LTE?

Even if some competitors will make 5G chipsets, they still will have to build in everything from good ol' GSM over EDGE over HSxPA to 4G in addition, because what use is a phone (or IoT gadget!) that only works in major urban areas where 5G deployments will happen first? All this will have to incorporate Qualcomm patents at the least, and most likely also Qualcomm chipsets.

Hell when I drive to the German country side I more often than not only have EDGE, and the situation in developing countries is even more dire.

Qualcom got working on 5G technology as soon as 4G was finalized. I'm not an expert but it seems that the standards are written to be very close to the technology that Qualcomm is developping in parallel, if only because they demonstrate feasibility. They have a tremendous amount of headstart and expertise, which makes it very hard for new actors to enter this market. Also the amount of overlapping patents between successive technology generations is probably a big deterrent as well.

5G is more like an evolved 4G. So nothing changed much. (Qualcomm still biggest Winner ).

percentage of royalties would reduce from current ~5% in LTE to 2.5-3% in 5G.many operators have just invested in upgrading to gigabit LTE so next rollout should happen after they recoup their current investments so probably by 2020.

It would be more surprising if they had accepted.

Negotiations 101, always reject the first offer?

I want them to merge just so the resulting company can be called BroadQual.

Nah, they'll probably "rebrand in order to harmonize on multiple vectors of international brand awareness and leverage strategic synergy among incipient blue-ocean markets" or some such, and then hire a bunch of high-priced consultants who will come up with something inspired like:


[1]: https://en.wikipedia.org/wiki/Borland#The_Inprise_years.2C_a...

Why not CommCom?

does HN hate humour? but I would call it Comm^2(square)


Qualcomm did the right thing.

Avago Corp. now called Broadcom run by that Monkey man Tan will ruin Qualcomm into bits and pieces. Just like how Broadcomm was ruined, back then the key people left & by selling tons of IP slashing $300M off the same fate will happen to Qcomm, first they will strip off the mega Patents which Qcomm holds and sell them off to offset this $103Bn. Because that Tan is interested in the NXP only due to the massive Automotive and IoT treasure (https://www.eetimes.com/author.asp?section_id=36&doc_id=1332...)

Surpisingly this is where Apple comes. Apple's iPhone's chipsets have tons of Broadcom tech and they share ties with them, Apple may even buy them, Slowly the Android will die because no other corp apart from Qcomm manages the OSS SRC development, CAF and XDA development speak for themselves. For the better or worse, Qcomm is the ONLY leader in the OSS SoC game. Period.

Because Intel doesn't do that and they retired after the Z3000 series powered ZenFone2 which has great processing at CPU with x86 for Android but poor SGX GPU from Imagination Technologies & total lack of drivers and documentation made no OEM to opt for Intel.

Exynos (Samsung's) focus shifted when past Exynos powered SII, SIII was the last. Now the cat and mouse game happens with their Knox lockdown on the Note5+ phones.

Mediatek is not even worth considering they are anti-GPL, HiSilicon (Huawei's) Kirin is in the same page as Exynos, Open Kirin is only a try to grab XDA attention and budget game. Because once you make the chipset with Neural Engine which will be under closed doors (NPU in Mate 10, Kirin 970) it won't be like Tensor Flow at all. Forget the OSS development under it.

Xiaomi doesn't bode well either with their super long 3-6mo wait times for kernel src drop they don't even stand a chance against the Qcomm's Snapdragon SoCs and their CAF in both performance or the Open nature. (Read- https://www.xda-developers.com/gplv2-and-its-infringement-by...)

Also Qcomm is responsible to destroy the SD80x platform off support due to the CTS check for vulkan driver and lack of Nougat (Android 7.x) Reason ? Because these Corporates don't want to trust the OEMs, and provide all documentation, If it leaks their IP is at stake of imitation, security issues. It's a dangerous game. (Read- https://www.xda-developers.com/in-depth-capitulation-of-why-...)

And also Pixel, it's way worse now with removal of Accesibility services threat (Read - https://www.xda-developers.com/google-threatening-removal-ac...) & (http://www.androidpolice.com/2017/11/12/google-will-remove-p...) (https://www.reddit.com/r/Android/comments/7c4go5/is_google_p...)

Removal of overlay over other apps (Read - https://www.xda-developers.com/android-o-is-breaking-apps-th...) Removal of granular battery stats (Read - http://www.androidpolice.com/2017/03/22/android-o-feature-sp...) their new Pixel's Visual core is made by Intel (Read - https://www.xda-developers.com/google-intel-pixel-visual-cor...) and the functions it offers so far on the 8.x DP is just 2 (Read-https://www.xda-developers.com/developers-cant-test-hdr-pixe...)

So basically Google is now becoming Apple #2, with slamming doors infront of developers, Headphone jack fiasco, Overpriced trashware iPixels, also they have a secure Element chip + A/B system making TWRP issues...(Read - https://www.xda-developers.com/magisk-v14-4-root-pixel-2-xl-...)

And coming back Qcomm did every innovation and acquired Patents legally for more than $25Bn, Most of them include 4G/LTE and 5G R&D having a lead over Intel/Mediatek by 20Months but they are raising price for that chip or license deals & as Apple wa earning a lot over the iPhones, Qcom license works per phone profit basis. So they started this war against Qcomm since they are not new to kill off the ones who have higher revenue as their producer-customer relationship like GT Adv tech ( Read - https://www.reuters.com/article/us-gt-advanced-tech-bankrupt...), Imagination Tech (Apple's so called inhouse GPU is nothing but Imagination's IP after poaching a few of the key people they decided to drop ball Read - https://www.anandtech.com/show/11243/apple-developing-custom...) have Intel,which are inferior to the X16/20 modems at just 600Mbps vs Qcomm's 1.2Gpbps downlink.

Apple vs Qualcomm started long back, dating back to 2007 and WiMax threat to Qualcomm, (Read this - https://www.eetimes.com/document.asp?doc_id=1332242) but now Qcomm gets razed due to the Apple's global domination. Plus NXP antitrust scrutiny. Unless they get that NXP, Apple will continue to blaze them & Broadcom too.

This is what Apple does to industry, I suspect wearing my Tin Foil hat, They pressed Broadcom of this deal on Qualcomm, now even Huawei is dropping royalities for Qualcomm and SKTC asking to renegotiate Qcomm IP licensing it's a mess, because despite the Qcomm's nature of Patent wars they are the only OE who support OSS which is most crucial aspect nowadays regarding a smartphone vs Planned obsolescence.

Ultimately Broadcom is dangerous. NXP deal should move else Qcomm will lose money due to Apple war and there are ton looming around..& remember vote with wallet.

103BN is more than a GDP of entire countries. This is just mind blowing.

You're confusing a stock and a flow. GDP is a measure of value-added per period. That's a flow. The value of a company is a stock. More comparable to the wealth of a country than its GDP.

First, that's not necessarily the case (that they're confused), the parent may simply be referring to that number for scale purposes.

Second, what they said is correct on an annual basis. The value of that merger, at $103 billion, exceeds the annual GDP of numerous nations.

Before you make an absolute statement about their confusion, perhaps you should ask for clarification first. What they said was not strictly wrong.

The economic output of an entire nation can be useful for such comparisons when you're attempting to relate to very large financial figures.

Sheesh.. pedantry on HN can be harsh. I remember sparking about 15 cranky comments for calling Milton Friedman “on the Austrian side of economics.”

But, can’t beat em, join em.

Qualcomm is basically Estonia. Qualcomm’s revenue multiple is about 4.25X. Using that multiple, Estonia should have a similar valuation.

Qualcomm has the higher growth rate, but Estonia has been a country for a while with a loyal customer base. Qualcomm’s customers will probably use whatever chips come wired into their phone. Estonia’s are likely to stay Estonian unless a disruptive innovator comes along with a 10X improvement. 10X, not 10%. Think big Putin. If we weigh these against each other, it’s about even.

So… Estonia’s National valuation is about $103.4, based on $24bn in gross stuff (yuck).

Just kidding!. That’s preposterous of course. Qualcomm is investor backed, Like Tesla. Estonia is debt backed, like a country. You can’t compare those either, and you’d know this if you read all the cranky HN pedantry from before! Someone tried to compare Tesla to GM, the fool!

What we need is Estonia’s enterprise value…. Now, that’s apple’s to apple’s. Also, Russia, NATO and stuff.

You simply said something that was flat out wrong, that does not people were 'cranky'.

That was a different individual than was replied to up thread.

I think parent meant this:

> calling Milton Friedman “on the Austrian side of economics.”

Which is pretty laughably wrong for someone at least a bit educated in the area.

You consider him more of a Menshevik?

I did not know that economics was all about Austrian vs Menshiviks.

That part is straightforward. Traditionally, economic dialectics has two shivics at a time.

But, since you've conceded the point on pedantry, I'll give you a 3rd shivic. Which would you prefer, chicagoshivic?

"The economic output of an entire nation can be useful for such comparisons when you're attempting to relate to very large financial figures."

I repeat: It's comparing a stock and a flow. You cannot compare a stock and a flow because they are different units. 103bn per year vs. 103bn as a volume of wealth. You could compare 103bn to the wealth of a country. Or you could say that it is equal to N empire state buildings, or a number of dollars that goes around the world X times. But if you compare it to GDP, then you're not comparing apples to apples.

The problem is that flows like GDP count the "same dollar" multiple times. When the dollar moves from one place to another, it's counted. (To preempt the pedants: Yes, there are exceptions as to exactly where these counters are. Not every movement is counted.)

That's drastically different from having to have a large pile of "different dollars" to make a large purchase like this.

GDP is a measure of value-added. That's why they net out things like net trade.

What is mind blowing is that even $103bn is not a lot of money in the grand scheme of things. $100m isn't even a rounding error in the global economy. AB Inbev, a beer company, also recently paid $100bn to acquire their main competitor.

That puts into perspective how reasonable it is to expect to be able to make $10-20-50m building and running a business.

AB InBev growth and mergers is astounding. The number of companies that form the [original base] of the company is insane.

To be fair to AB InBev, they had to give up substantial portiona of SABMiller to get the merger through which as you said was a $100B+ [cash] deal. Combined company revenue is supposed to rise by around $11B this year. While SABMiller did $19B in revenue alone before the merger. SABMiller sold off at least $25B of the company before or during the merger. Not always for the best price either as all sides knew the sales had to happen.

Honestly, I think Inbev is one of the most interesting cases when it comes to wealth building. Interbrew was a merger of small(er) Belgian breweries, run mainly by old money aristocratic families. And they basically levered and merged their way up, into a global player worth hundreds of billions in market cap.

Definitely. It's like a handful of Belgian families and the Brazilians at 3G owning a majority of the company still after all this time. Amazing.

I don’t see at all how it makes it reasonable that a person should earn that much running a business at all. Maybe it’s reasonable to the board of that company but I don’t consider it reasonable.

I think you interpreted the comment wrong. Of course it's not reasonable for them to expect that money to just show up. But it's perfectly reasonable for them to expect that they could capture that much business. Because there's a lot of money flowing all the time, and 50 million is really a very small portion of it.

GDP is a measure of everything produced in a country per year. It makes no sense to compare the value of a company with growth per year.

Of course you can compare a rate with a quantity. GP is saying that whole countries take more than a year to create one (proposed) Qualcomm’s worth of value.

Have you never compared your car’s gallon-per-mile rate of fuel consumption with its gas tank’s size in gallons?

I haven't. The operation I suspect you're thinking of is division, not comparison. Division of capacity by consumption gives range, whilst a comparison leaves me scratching my head for a meaningful interpretation. I'm afraid this analogy appears to reinforce the point rather than challenge it.

But by that logic, it makes no sense to compare the cost of a new computer to the cost of rent, because they're different things. GDP is a measure in units of money, and so is stock value, so you can compare them just fine even though they're not related to each other.

That doesn't follow at all... A cost is always an outflow of money. As I mentioned in another thread, GDP counts the "same dollar" being spent multiple times. Even if there was only one dollar in the economy, you could generate millions in GDP by spending that one dollar millions of times in a year. However, you sure as hell aren't going to be making a $10 acquisition proposal with only $1 in the economy.

You're missing the point here, no one is comparing the two as if a country's GDP represents something equivalent to the value of a company. They're just saying something along the lines of "Wow! 3,000,000 gallons of water?!? That's more than the amount of water that flows past my house on the river in a year!" (Obviously made up numbers)

Well if you're going to try to be technical about it, GDP does not represent growth per year. GDP also does not specify that it must refer to a year of time, you have to qualify that. GDP is also frequently measured quarterly.

It makes sense to relate to GDP if what you're looking for is a well understood scale to use as a reference for very large financial numbers.

Since the GP did not mention a time frame I assumed yearly (otherwise it would be most countries instead of some). And when you compare GDP with a the value of a company it sounds like “company X is more valuable than country Y” to most people. I was just trying to mention that this is not the case.

I thought Qualcomm had better revenues?

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact