Comment by ggm | Hacker News Reader

Comment by ggm | Hacker News Reader

Parent Full thread

Decade or more ago, Apple flagged they were heading to in-housing the profits in VLSI when they cut ARM out of the deal behind the chips for the iPod. Nobody should be surprised if somebody with big money in TSMC sees Apple (who buy a lot of their product) as a better investment. (they kept the IPR tie-in within limits but stopped using them to be the foundry and started adding their own FPGA and the like to arm cores. So it was functionally diluting)

TSMC are going to be exposed to high risk re-building in the USA (making VLSI plants is very very expensive and yield can be many years in the hunting), and have bottlenecks (water access) in Taiwan as well as the war risk. Maybe Berkshire sees them as a dropping value shere for a bit. I doubt they won't return, maybe later on when TSMC shares are better value? Berkshire is a hold company, not a trade-at-a-rate-of-knots company.

I'm not sure why Berkshire thinks banks are a bad bet: they have made bank (sorry) this last downturn, and will continue to. A lot of profit in them yet, and with interest rates going high, people will be flocking to them for income generating products like annuities. Maybe I misunderstand the risk here but I would think a bank is a better bet right now than listed property trusts.

If I am not mistaken, Berkshire is huge in re-insurance and in general goods and services. If they move out of the insurance market and cite global warming risk, thats the signal everyone needs to see that its a real risk, not some fancy.

Is Berkshire invested in the dutch chip-making tool maker side of things? Thats got to be the shovel-seller in the goldrush.

[edit: I am hazy on exactly what is the role between Apple, ARM-UK holdings, Wolfson, TSMC and their history of chip design so please don't beat me up if you know ARM was never the foundry. The point was, Apple used UK VLSI designs, secured designs, and then moved into the territory to internalise profit. exactly what they did I don't know but I do know they marginalised work that was done for them]

Insurance is a core tenet of Buffett’s investing philosophy. He uses the “float” from premiums to plow into other investments and then he has the entire cash flow juggernaut of Berkshire to cope in the years when insurance has claims loss. And he’s gotten so amazing at it that he’s able to underwrite risky/exotic policies and profit handsomely from it.

Climate change is definitely a hot topic with insurers but it’s not going to cause them to go broke.

Geico is such a cash cow because unlike most insurance businesses, there is vary little variance. The average claim is low and it's smoothed out by having so many customers. There are no large ,discrete events. All it needs to do is raise premiums to ensure large annual , recurring profits no matter what.

Aren’t the reinsurers on the hook for the earthquake in Turkey?

Someone maybe, and maybe they'll pay if they want to keep working in Turkey, but... many of the builders were given amnesty for violating construction rules, and then arrested after the earthquake. There's video of Erdogan boasting about how much more housing was put in from the amnesty. So maybe Reinsurers claim most of those buildings aren't covered, because of building violations? That's what they'd do in Florida.

https://www.npr.org/2023/02/13/1156512284/turkey-earthquake-...

I don’t know the rules in Turkey but in the Netherlands insurers cannot insure major disasters like earthquakes, war and nuclear explosions, because although the events are extremely rare a single one would bankrupt the companies.

They’ll just refuse to cover the cases when it would, easy peasy.

Thats kind-of what I mean: if Berkshire "invites" its RE investments to cease laying off climate change risks, or to make immense profit for the risk, then the worldwide re-insurance market is going to move, and everyone holding waterfront property is going to find premiums jump no matter what because Re-insurance spreads risk: everyone winds up paying for that spread, down at the coal face.

"oh sorry, the cost of re-insurance jumped 10x, your premiums have to go up" is a hell of a signal.

See Florida property insurance.

Nationalization of industry will likely rise with deglobalization, thats pretty risky for international banks.

Do you mean localization?

"Nationalizing an industry" means a country's government coming to own and operate said industry. See, for example, railroads of ye olde days.

I mean nationalization. As the world economy fractures and countries take more control over their supply chains nationalization of industry will take place. China does it all the time. Russia is planning on doing it. The US is talking about doing it (and has to some extent) to combat China's influence domestically.

It's going to increase dramatically all over the world, deglobalization means supply chains are a security concern as much as if not more than an economic concern now.

> yield can be many years in the hunting

Not really. Semiconductor yields are highly correlated between different sites [1].

[1] https://en.wikichip.org/wiki/intel/copy_exactly!

^ HN strips the ! from the link, so please add it at the end of the URL.

> ^ HN strips the ! from the link, so please add it at the end of the URL.

That's because it's not a valid URL char, just browsers do the translation for you. You'll want to URL-encode it. Replace `!` with its url-encoding %21:

https://en.wikichip.org/wiki/intel/copy_exactly%21

They see the conflict approaching faster than anticipated, leaving not enough time for the US to delocalize TSMC

I suspect we'll see a TSMC sabotage in the coming months

https://www.theregister.com/2022/01/05/taiwan_should_destory...

We have that, and also Huawei's EUV patent, Apple perhaps has something hidden yet to be announced

https://www.techspot.com/news/97072-huawei-patents-euv-litho...

> I suspect we'll see a TSMC sabotage in the coming months

What evidence exists to show that China is preparing to annex Taiwan? Satellite troop movements? Natural resource hoarding? Anything at all?

If China, or anyone, invaded Taiwan it would be fantastically easy to cripple tsmc fabs. Kill their engineers en masse. Blow up the lithography machines. Unleash a targeted virus like stuxnet designed for the hardware in the facilities on their networks. An important question remains, does China have the footing to not only fill the vacuum in their supply chain but to also weather the sanctions and quite possibly engage the us and its allies without turning cities to irradiated glass and starting nuclear winter?

I seriously doubt that China is secretly prepared to meet their own domestic demand for advanced silicon let alone call bluff of the five eyes.

That said, maybe xi is actually more out if touch than Putin and we’ll be cowering in fallout shelters come summer.

I don't think they'll "invade", a word that's mostly used here in the west, mostly in the US, China doesn't seems to want a war, Taiwan even less, its population rejected the pro-US government [1]

On the other hand, the US proved that they are willing to replace the goverment of a country and arm it

> I seriously doubt that China is secretly prepared to meet their own domestic demand for advanced silicon let alone call bluff of the five eyes.

That's the same people who; 10 years ago said "i don't think China will be able to put into orbit a space station, let alone visiting the moon"

[1] - https://www.dw.com/en/taiwan-president-resigns-as-party-head...

IMO TSMC doesn't really play much a role in the Taiwan issue. It's all about geopolitics. Taiwan is right off the east coast of China, and also right in the midst of all major East Asian sea trade routes. And the US is continuing to ply them with tens of billions of dollars in weapons and overtly looking to turn it into a defacto launching pad for US forces. That's a rather severe threat from China's perspective, regardless of whether Taiwan was a manufacturing center of the world, or just another Guam.

Sure but that's been true for decades. The question is, what's new? Here in Taiwan it's business as usual. Twenty years ago, "idk about buying property, what if the PRC invades?" Ten years ago, "what if the PRC invades?" Meanwhile earthquakes, tsunamis, hurricanes, etc, causing disruptions here in Taiwan and everywhere else in the world.

So what's changed today that makes a PRC invasion more likely than a major earthquake, which we already just live in peace with the possibility of?

I'm going to avoid answering this, largely because it's impossible to do so succinctly. Obviously a massive amount has changed as I'm sure you know, too much to cover succinctly. And then there are also the subjective factors. As but one example, how one thinks the war in Ukraine is going, for instance, is going to inform their opinion on China's self-perceived relative strength/weakness.

Ultimately, if you believe a conflict is probable, it comes down to a question of timing. China will want to make a move when the US is at a weak point and/or distracted with other obligations. Similarly, the less favorable (to China) maintaining the status quo is, the more inclined they will be to act.

> Obviously a massive amount has changed as I'm sure you know, too much to cover succinctly.

This is exactly why I put my foot down and challenge hard whenever I hear, "oh, well, you know, the China situation in Taiwan."

No, I don't know. And nobody can tell me. Why is that? All I get is hand waving.

I'm not trying to be reductionist or asking people to oversimplify, but I do think it should be possible to answer the question of "what indicators of invasion are strong enough to prevent investment in Taiwan companies" with something succinct. Typically people go the route of "increased air traffic from PLA jets" or something, though that's historically consistent and also hasn't resulted in invasion in 3 decades. Not even local regime change caused an invasion when Taiwan switched to democratic rule.

> subjective factors

I argue that these "subjective factors" are really just artifacts of a media system that prefers to distort reality as a far more interesting and violent place than it really is. I argue that instead of Taiwan being a very exciting, about-to-be-invaded-any-minute-now place, it's actually a very mundane place, facing the same catastrophic risks of any other mundane place, such as Miami or San Francisco (or both I guess, since Taiwan faces both typhoon and earthquake risk. Still pretty mundane compared to how the media portrays us).

I enjoy gametheorying as much as anyone else, but I'm noticing that this vague handwavy subjective "invasion risk" is having real effects, and that annoys me. I've had people say, "I'd visit, but I'm worried about the China situation." What? Really? But you'll get in a car and face down unknown numbers of drunk drivers every day? I don't get it and I'll be honest, it annoys me a bit. So, hence why I push back so hard.

I don't think anybody's claiming Taiwan is violent or especially "interesting." I spent a while there myself and quite enjoyed it - the best public transport I've experienced anywhere in the world, completely bemusing escalator behavior, sewer smell tofu and all. But the nasty thing about war is it's not like there's some visible meter that fills up, and once it reaches the top - war starts.

It starts when one side believes they have a strategic advantage, and often it comes exceptionally fast. Ask most people who are familiar with the situation in Russia/Ukraine, including Ukrainians and Russians, whether they thought Russia would invade in 2022, even just days before it happened, and most would have laughed at the idea of it. Even Zelensky was [publicly] overtly rejecting the US claims of an imminent invasion.

With Taiwan both sides are being increasingly provocative. There was absolutely zero purpose in sending Pelosi to Taiwan. It was essentially just daring China to stop her. It's somewhat similar to how in the leadup to the war in Ukraine numerous Western vessels were semi-regularly violating Russia's territorial waters (based on Crimea), essentially daring them to escalate. The CCP, for their part, have also stepped up their rhetoric and there were even leaked (more probably "leaked") plans for an invasion of Taiwan. Even if disinformation, it's obviously designed to provoke. Basically both sides are poking and prodding one another, to feel each other out. I don't think it's simply posturing.

There are also somewhat practical indicators. China has been accelerating an ongoing process of de-dollarizing itself, engaging in simulations of 'Russia like' sanctions on its economy, stockpiling food [ostensibly for COVID lockdowns], expanding their nuclear arsenal, increasing conventional arms production [ostensibly for export], and more. And while I'm skipping most of the geopolitics stuff for brevity, I would at least mention the obvious indicators that the Mideast (Saudi Arabia and UAE in particular) seem to be moving hard towards the 'Eastern bloc.' This sort of an alliance opens up doors that are big enough to start speaking hyperbolically in terms of outcomes.

Correct me if I'm wrong but retail banks historically aren't cash cows, they return about 1% of assets on deposit. They generally have to pass on the better interest rates to their customers. In a world where the risk-free rate is like 5%, it probably makes sense to divest of banks and look for higher yields in something like Apple. Their stock pump may have just been reflective of transient economics. Just speculating, this isn't my area.

[edit] Also, personal savings rates are crashing in the US and revolving debt utilization is way up. Maybe Berkshire is worried about cyclical defaults and withdrawals of principal? Even if bank ROA is up, lower deposits mean lower revenues. [1]

[1] https://fred.stlouisfed.org/graph/?g=VtLo

> return about 1% of assets on deposit

So? Investors don't have to lower-bound value bank stock at its deposits.

(God I hate google and its insistence on corporate search results, I search for "how much money does bank of america have on deposit" and all the results are about making deposits on bankofamerica.com)

Bank of America has $1.9t in deposits, but its equity value is $285b. So a 1% return on deposits = 6.66% return for investors (excluding other costs and other revenue (account fees?))

That's odd. When I type your query into Google, I get a table at the top with:

    Rank  Name              Total Deposits
    2     Bank of America   $1,906,458,000
    3     Wells Fargo Bank  $1,479,499,000
    4     Citibank          $1,282,071,000

And a link to https://www.mx.com/blog/biggest-us-banks-by-deposits/

I get the same result whether I'm in incognito mode or not. Are you perhaps running an adblocker that's removing this?

typing the same query I get a full page of bankofamerica.com results, even after disabling the blocker.

Adblocker, yes, but it wouldn’t block that. I’m not in USA though.

Side note re: search, Bing Chat answers your question (cut and paste from your quote):

Bank of America has different types of deposits, such as interest-bearing deposits, non-interest-bearing deposits, and mobile deposits. According to the annual Summary of Deposits from the FDIC, Bank of America had $1.36 trillion in total deposits as of June 2019(1), making it the largest US bank by deposits. However, this number may have changed since then. The Bank of America mobile check deposit limit is $10,000 per month for accounts opened for three months or longer, and $2,500 per month for accounts opened for fewer than three months(2).

Learn more: 1. retailbankerinternational.com 2. gobankingrates.com

What a strange mix of corporate financials and retail product specs. The first sentence implies that “mobile deposits” are of a kind with interest- and non-interest bearing deposits. The last sentence talk about the process of making a deposit into your own account.

The middle two sentences actually answer the question regarding the total volume of deposits across the bank.

Verdict: Still useful! Non sequitor-ish, but better than SEO spam :)

Or is it? The Bing answer is about $600 billion less than the human generated answers.

One of them is wrong (or at least outdated), I don't know enough about BoA to know which though.

Indeed. Or, as Bing notes, is the definition different? There is a sourcing link (my cut paste of the HTML did not get the full URL, unfortunately)

Aside: I would expect equity value of a bank to be predominantly dependent on returns, since many banks don’t have a lot of assets (except perhaps investment assets that produce returns and the assets themselves are valued on their returns).

> equity value of a bank to be predominantly dependent on returns

bingo. Though you have to worry about leverage (ie: risk). Really sucks when a small change results in catastrophic losses (though everyone loves it on the way up).

Berkshire has a considerable amount of Bank of America stock, whose profit margins are trending very well:

https://www.macrotrends.net/stocks/charts/BAC/bank-of-americ...

Its other big bank holdings, Amex, Citi, Bank of NY Mellon, and US Bank also have healthy double digit profit margins.

https://www.cnbc.com/berkshire-hathaway-portfolio/

If you hold enough deposits, 1% is pretty good. Remember, these assets aren't "your" assets. you hold shares in the bank. the 1% isn't a rate of return on the share pool. Banks pay dividends reliably.

Reply on news.ycombinator.com