5 ms·

They have so much cash, and the business is so valuable there isn’t a better place to put it.* So you give it back to the shareholders, and a buyback is tax ad

by gumby 2y ago

They have so much cash, and the business is so valuable there isn’t a better place to put it.* So you give it back to the shareholders, and a buyback is tax advantaged for the shareholders compared to a dividend. * if you are, say, Pepsi, and you have a very good year, you might want to “bank” some or all of those profits for a lean year or for some future investment. But the CFO doesn’t put $10B in a bank account, they invest it. If they put 10B into Apple it won’t have any legal implications (Apple is so huge) and it’s probably a better return than anything else available. But when you are Apple the sums are so huge and the choice of places you can invest so limited, that giving it back to the shareholders and letting them decide is pretty much your only option. A good problem to have.

JKCalhoun 2y ago
That kind of money ... they could build their own fab? Other than that, maybe R&D? Even if Apple were known as an R&D company (and they haven't been known for that since Jobs returned and among other things got rid of ATG), probably wouldn't even need that kind of cash to run.
- dtech 2y ago
  I'm sure they do a lot of R&D. They built their own CPU for goodness sake. But there's diminishing returns, at some point the ROI for the next candidate investment isn't good enough to make it a valid choice over a buyback or dividend.
  - alt227 2y ago
    Minor nitpick - they didnt build their own CPU. They licensed a design from ARM, tweaked it, and sent it to TSMC to manufacture it.
    - gumby 2y ago
      They have an architecture license, so can do quite a bit beyond simply "tweaking" it. They have some degree, presumably a significant degree, of "blank sheet" implementation. At this point, probably the main benefit of the ARM license is just the patents.
      - alt227 2y ago
        I have read that the architectural license comes with access to ARM engineers expertise, internal build tooling, validation and testing systems etc. So whatever processor design Apple may be doing, they are definitely doing it with the guard rails on.
        gumby 2y ago
        Those few architectural licenses are secret, but surely, as you say, allows Apple to discuss all sorts of things with ARM’s engineers. I have long wondered what constraints it might carry. They have added an instruction or two, but I doubt there is any real reason for them to ever want to make significant instruction set changes. Apple has a longstanding relationship with ARM back to the latter’s founding. The sale of its large chunk of ownership in ARM helped Apple survive in its late-90s reboot. So it’s quite possible that their license is quite different from those of the few other hardware licensees. Also, Apple plays the long game. I’m sure they also have a lot of RISC-V experience by now. Not just weirdo in-house research products; I’d be shocked if there weren’t tiny RISC-V cores doing small housekeeping tasks inside some of their huge chips, just for the experience. Maybe even a few inside an Apple pencil or maybe an ipad screen digitizer since the computation isn’t really user visible. As does ARM: if RISC-V ever becomes airborne (it’s still speeding down the runway) I suspect ARM will haul out a stack of patents and start demanding payment. Apple’s license surely covers them here.
- rTX5CMRXIfFG 2y ago
  Corporate diversification entails risks and fails 70-90% of the time.
- Tuna-Fish 2y ago
  Could they expect to do better than TSMC with it? Seriously though, people decrying this as financial engineering seem, to me, to be completely missing the point. It's not the purpose of a firm to endlessly expand their business until they do everything. It's to do the thing they are good at, what they have a comparative advantage over the rest of the industry over, and then instead of chasing ever worse roi on ever more marginal investment and research, return the rest of the money to the investors so they can make their own choices on what to invest on.
- airstrike 2y ago
  Companies have a limited number of R&D projects and not all of them have great expected returns on investment, so sometimes it makes more sense to wait before investing in Project n+1
  - JKCalhoun 2y ago
    To be sure. There is a popular sentiment that the Next Big Thing is just within reach, is always within reach, if the big guys like Apple or Meta would just throw enough resources at it it would come to exist. I think that is fallacious thinking. Apple tried hard with the Newton but larger, infrastructure pieces were not yet in place. Meta has tried hard with VR but there is no evidence that a large population even wants this tech. Apart from AI, we may simply be in an innovation desert for the consumer tech companies. Things like the portable digital music player, smart phone, might only come every decade or so — and no amount of R&D will force that to behave any differently.
- gigatexal 2y ago
  Hmm. 200M phones a year and ~50M computers and some number for million of accessories if the Fab could do the whole gamut and do it at as-good or better than TSMC scale hmmm I doubt it would be a profitable endeavor in the long run
Zenzero 2y ago
> But when you are Apple the sums are so huge and the choice of places you can invest so limited, that giving it back to the shareholders and letting them decide is pretty much your only option. A good problem to have. Crazy thought but I know one larger entity to park the money in and it's the US treasury. Apple has no positioning to improve things like infrastructure and public education. Stock buybacks should be a clear signal that taxes can go up at that bracket of the economy.
airstrike 2y ago
> If they put 10B into Apple it won’t have any legal implications (Apple is so huge) and it’s probably a better return than anything else available. While this is true, investors would generally rather have the money returned to them so they may diversify their portfolios as they see fit instead of letting the company do it for them. (at least for spherical cow investors)
- appplication 2y ago
  Given the stock jump on the news I think these cows might be a bit less spherical than assumed.

by gumby 2y ago

JKCalhoun 2y ago

That kind of money ... they could build their own fab? Other than that, maybe R&D? Even if Apple were known as an R&D company (and they haven't been known for that since Jobs returned and among other things got rid of ATG), probably wouldn't even need that kind of cash to run.

dtech 2y ago
I'm sure they do a lot of R&D. They built their own CPU for goodness sake. But there's diminishing returns, at some point the ROI for the next candidate investment isn't good enough to make it a valid choice over a buyback or dividend.
- alt227 2y ago
  Minor nitpick - they didnt build their own CPU. They licensed a design from ARM, tweaked it, and sent it to TSMC to manufacture it.
  - gumby 2y ago
    They have an architecture license, so can do quite a bit beyond simply "tweaking" it. They have some degree, presumably a significant degree, of "blank sheet" implementation. At this point, probably the main benefit of the ARM license is just the patents.
    - alt227 2y ago
      I have read that the architectural license comes with access to ARM engineers expertise, internal build tooling, validation and testing systems etc. So whatever processor design Apple may be doing, they are definitely doing it with the guard rails on.
      - gumby 2y ago
        Those few architectural licenses are secret, but surely, as you say, allows Apple to discuss all sorts of things with ARM’s engineers. I have long wondered what constraints it might carry. They have added an instruction or two, but I doubt there is any real reason for them to ever want to make significant instruction set changes. Apple has a longstanding relationship with ARM back to the latter’s founding. The sale of its large chunk of ownership in ARM helped Apple survive in its late-90s reboot. So it’s quite possible that their license is quite different from those of the few other hardware licensees. Also, Apple plays the long game. I’m sure they also have a lot of RISC-V experience by now. Not just weirdo in-house research products; I’d be shocked if there weren’t tiny RISC-V cores doing small housekeeping tasks inside some of their huge chips, just for the experience. Maybe even a few inside an Apple pencil or maybe an ipad screen digitizer since the computation isn’t really user visible. As does ARM: if RISC-V ever becomes airborne (it’s still speeding down the runway) I suspect ARM will haul out a stack of patents and start demanding payment. Apple’s license surely covers them here.
rTX5CMRXIfFG 2y ago
Corporate diversification entails risks and fails 70-90% of the time.
Tuna-Fish 2y ago
Could they expect to do better than TSMC with it? Seriously though, people decrying this as financial engineering seem, to me, to be completely missing the point. It's not the purpose of a firm to endlessly expand their business until they do everything. It's to do the thing they are good at, what they have a comparative advantage over the rest of the industry over, and then instead of chasing ever worse roi on ever more marginal investment and research, return the rest of the money to the investors so they can make their own choices on what to invest on.
airstrike 2y ago
Companies have a limited number of R&D projects and not all of them have great expected returns on investment, so sometimes it makes more sense to wait before investing in Project n+1
- JKCalhoun 2y ago
  To be sure. There is a popular sentiment that the Next Big Thing is just within reach, is always within reach, if the big guys like Apple or Meta would just throw enough resources at it it would come to exist. I think that is fallacious thinking. Apple tried hard with the Newton but larger, infrastructure pieces were not yet in place. Meta has tried hard with VR but there is no evidence that a large population even wants this tech. Apart from AI, we may simply be in an innovation desert for the consumer tech companies. Things like the portable digital music player, smart phone, might only come every decade or so — and no amount of R&D will force that to behave any differently.
gigatexal 2y ago
Hmm. 200M phones a year and ~50M computers and some number for million of accessories if the Fab could do the whole gamut and do it at as-good or better than TSMC scale hmmm I doubt it would be a profitable endeavor in the long run

Zenzero 2y ago

> But when you are Apple the sums are so huge and the choice of places you can invest so limited, that giving it back to the shareholders and letting them decide is pretty much your only option. A good problem to have. Crazy thought but I know one larger entity to park the money in and it's the US treasury. Apple has no positioning to improve things like infrastructure and public education. Stock buybacks should be a clear signal that taxes can go up at that bracket of the economy.

airstrike 2y ago

> If they put 10B into Apple it won’t have any legal implications (Apple is so huge) and it’s probably a better return than anything else available. While this is true, investors would generally rather have the money returned to them so they may diversify their portfolios as they see fit instead of letting the company do it for them. (at least for spherical cow investors)

appplication 2y ago
Given the stock jump on the news I think these cows might be a bit less spherical than assumed.