Hacker News new | past | comments | ask | show | jobs | submit login
Intel ousts its chief engineer, shakes up technical group after delays (reuters.com)
559 points by myrandomcomment 12 months ago | hide | past | favorite | 519 comments

> Ann Kelleher, a 24-year Intel veteran, will lead development of 7-nanometer and 5-nanometer chip technology processes. Last week, the company had said the smaller, faster 7-nanometer chipmaking technology was six months behind schedule and it would have to rely more on outside chipmakers to keep its products competitive.

I hope Intel keeps its US-based fabs and continues to upgrade them. It would be a mistake to cede this capacity to outsourcing like so much of our other manufacturing capacity. Interestingly, it seems the US is almost unique in our mass deindustrialization. Japan, Germany and South Korea have avoided this.

This deindustrialization was driven by Wall Street. The important metric is return on invested capital.

Manufacturing companies use a lot of capital for their factories. Shut down the factory and outsource to China, and you have the same sales with a lot less capital, so the stock goes up. Great success.

Maybe repeat by having your suppliers do more and more of the engineering. At a certain point, you are just a brand, with no actual engineering capability.

You can't compete with the people who actually know how to make the things. Then you die. See Hewlett Packard.

The US taxing companies on global income at high rates also has an effect. Say you have 50% of your sales outside the US. You pay taxes in the country where you sell the product. Then you have profits overseas. If you bring that cash back to the US, you pay the difference between the corporate rate in country X and the US, which can be a lot, say 15%. Might as well take that money and use it to pay for manufacturing at e.g. Foxconn. So it's 15% better to invest in manufacturing capacity outside the US.

The poor health care system in the US drives up costs as well, twice what it is in other industrialized countries for worse outcome. Same for cost of housing because zoning laws won't allow building. Google was allowed to build offices in Mountain View but not residential property for the employees to live in. So everyone has to commute from somewhere else, and cost of rent is driven by how much misery you can handle.

Failures in government policy make US workers much more expensive without creating any more value or improving the quality of life for the employees.

> You can't compete with the people who actually know how to make the things. Then you die. See Hewlett Packard.

Innovation happens on the factory floor. Pretty soon the countries we outsourced to will come up with better designs too. The U.S. thesis for outsourcing is that the manufacturing countries are filled with braindead automatons who can't compete with our "Designed in California". That may have been true in the 20th century when the U.S. brain drained all the top talent, but now other nations are in a position to pay their engineers more than U.S. companies. The "Designed in California" cope can only last so long.

As someone that has worked for several years in this area with some major Asian offshoring IT shops, I have seen that mentality myself, although from European point of view.

People that reach for this kind of decisions occasionally seem to still have a colonialism point of view, like "they will do as told and we still stay in control", except that isn't how things work and eventually they stand on their own feet and are able to even outperform their "masters".

> People that reach for this kind of decisions occasionally seem to still have a colonialism point of view, like "they will do as told and we still stay in control", except that isn't how things work and eventually they stand on their own feet and are able to even outperform their "masters".

Oh man, the amount of Dutch "we're going to build a knowledge economy" news articles I've seen in the last decade that barely hide their colonial mindset is staggering, and the implied assumption that "we" are smarter than "them" is so blatantly racist that I'd almost feel offended if I didn't know that they'll reap what they sow with that attitude soon enough.

It's not just Dutch but been a prevailing tune in most of the Western Europe (outside perhaps Germany) for a couple decades now.

And yes it stinks of "we'll live with the hard part of telling other people what to do".

>outside perhaps Germany

You clearly haven't dealt with German managers at their companies operating in Eastern Europe ;)

Technically that makes it a German company outside of Germany :p

Jokes aside, "behaving badly abroad while showing their best behavior at home" is basically yet another proud colonial tradition, isn't it?

Like how the Swiss are considered the most civilized people in Europe ... until they go driving in Germany or go partying in Mallorca/Prague.

I can’t count the times I was greeted Maestro! by the typical blond middle-aged manager from Amsterdam... (I’m Italian, so I guess we’re all Pavarotti even without the waistline.) I’ve been toiling 10years in NL and I’ve yet to see anyone in the executive suite who’s not a Heineken buddy from the Universiteit

> I’ve yet to see anyone in the executive suite who’s not a Heineken buddy from the Universiteit

I wish this didn't ring so true. It is so prevalent that for those who were raised locally and who went through said university schooling it is still painfully easy to distinguish from which student union in which city these guys originate based on how they treat each other, even when they graduated decades ago. Which ironically means that they are guilty of the kind of primitive tribalistic nonsense that they project onto other cultures.

... actually, now that I wrote that out, I guess that could partially explain a few things as well.

(and I'm well aware that this is not a problem unique to the Netherlands, but it still sucks)

Thanks for pointing this out.

I was working for a major Dutch electronics company and it was interesting to watch how whenever colleagues from their US/German brach would come visit the Dutch managers would treat them like equals and invite them to drinks and dinners but whenever colleagues from their Czech/China/Singapore branch would visit, the Dutch managers would politely pretend they don't exist.

This is so sad. Would you say the same treatment was given by the lower-downs?

>I can’t count the times I was greeted Maestro!

As an Eastern European immigrant, I can think of worse things you can be called in the West than Maestro!. You have no idea how good you have it.

Happens to me a lot too, but there is nothing remotely Italian ... or Dutch about me. I think they use it as a term of endearment to talented colleagues?

>I was greeted Maestro!

No Problem with that ;)

Out of curiosity, could you share some of them?

"People that reach for this kind of decisions occasionally seem to still have a colonialism point of view, like 'they will do as told and we still stay in control', "

It's not necessary colonialist as such, merely takes the past history as implications of future patterns. The so called colonialist powers were ahead in the game from 1700's up to the middle of the 20th century.

Colonialism coincided with the rise of industrialization and science. The so called "colonized" nations did not have the chance to utilize these and rise up to their full human potential.

I think a good historical example is that of Japan. Japan was effectively a medieval state in 1850 when the Perry expedition forced it to the modern era [0]. And 50 years later they beat a major world power in a technological military engagement in the Battle of Tsushima [1] against Russia! Gaining 500 years of advancements in technology in 50 years really amazing. But it really just reflects that we are all human, and all successes are due standing in the shoulder of past giants. And late comers to the game still have those same shoulder to stand on.

I suppose, then, half a century - or, two generations - is a good rule of thumb on how long it takes for a sovereign power to play catchup and to leapfrog to current "state of the art" just as long as it has a functioning state.

[0] https://en.wikipedia.org/wiki/Perry_Expedition

[1] https://en.wikipedia.org/wiki/Battle_of_Tsushima

The historical example of Japan absorbing knowledge and technology, leap-frogging to the modern era, is a good support for your argument.

There are fascinating artifacts from the period following Japan's opening up to foreign trade, reminiscent of European science in the Renaissance era - with an Asian twist.

I'd like to point out though, that there was a steady "leak" of technology, or rather cultural exchange with the West, through the Dutch since the 17th century.

Japan–Netherlands relations - https://en.wikipedia.org/wiki/Japan%E2%80%93Netherlands_rela...


To return to the "leap-frogging to the modern era" - it seems that developing nations are all in the process of doing this. I recall reading how many African countries "skipped over" the building of traditional telephone infrastructure, and went staight to mobile.

That also supports your argument that it only takes a couple generations for a country to catch up to modern standard of technology.

But to surpass it, and become the leading edge, probably requires a bigger cultural change - to have an academic and business environment that fosters innovation. In that aspect, I think the U.S. has a lead in the world, and will continue to keep that competitive advantage for a while.

It suggests that cultural (and perhaps social and psychological) change is harder to achieve than the knowledge transfer of technology.

Don't forget that even before the Dutch have shown up, there was portuguese missionaries in Japan, that among other things, introduced firearms there, that caught on fairly quickly because how easier it was to manufacture cheap firearms and give them to Ashigaru compared to training a full sword or yari samurai.

In a way, Japanese was ahead of Europe, using mass produced firearms in warfare earlier than Europe did.

Oh, that’s a good point!

However, after 17th century gun industry declined in Japan, and Japanese gun manufacturing was way behind europe in the beginning of 19th century [0].

But they really picked up the pace after that! Japanese Arisaka bolt-action rifles were the cheap-and-plentiful firearm used in many countries up to ww2. I have one Type-38 from 1905 or so and it sure is still very nice.[1]

[0] https://en.wikipedia.org/wiki/Firearms_of_Japan

[1] https://en.wikipedia.org/wiki/Arisaka

Ah, you're right - I forgot that the Portuguese had earlier exchange with Japan, during the 16th century.

Japan-Portugal History - https://en.wikipedia.org/wiki/Japan%E2%80%93Portugal_relatio...

Right, I'm connecting the dots, picturing the historical context..

That reminds me how Japanese children are taught in basic school about 黒船, "black ships", as representing first contact with the West.

Black Ships - https://en.wikipedia.org/wiki/Black_Ships

This was part of the Age of Discovery, as the Portuguese (chronologically) led the discovery of Africa, Asia, the Americas..

Supposedly tempura started as a result of Portuguese traders in Japan.

When it applies to international business, it can be read as colonial extraction, sure, but I think this mentality is a core part of what's taught in business schools for the management of domestic workers as well. This is the basis of Frederick Taylor's "scientific management" and the raison d'etre of the modern corporate structure - lots of upper management whose job is to be the thinkers on behalf of the doers.

Different countries have different corporate cultures. Differences in corporate culture are partially caused by differences in national culture. And, although I acknowledge it is a sensitive topic, I think differences in culture between countries can lead to differences in economic competitiveness.

The two countries I know the most about are my own (Australia), and the US. And I definitely see a lot of differences in corporate culture (willingness to take risk, thirst for innovation, etc) between the two countries, differences which generally come out in the US's favour.

I remember Elon Musk saying that "the United States is the greatest country that's ever existed on earth". The US has lots of quite serious problems and in saying that Musk is choosing to ignore all those problems. (And at least some of the problems the US has are caused by distinctive aspects of US culture). But, look at SpaceX – probably in no other country on earth could SpaceX exist.

Certainly I don't think anything like SpaceX could exist in Australia. Of course, the US has the advantage of 13 times more people. But, if you could imagine a country with the same population as the US but the same culture as Australia, I don't think anything like SpaceX could exist in that hypothetical country either.

If Musk's dreams come true, SpaceX could turn out to be the most important company in human history thus far. (Of course, maybe his dreams won't come true – but I hope they do – I would love to live to see the day that humanity becomes a multi-planetary species.)

> willingness to take risk, thirst for innovation

They are basically the same and it is a result of different attitudes of capital providers. In Australia, you have more conservative financial institutions and no venture capitalists -> so companies must take less risk.

Related to risk is also innovation. If you can take more risks, you are more willing to purchase product and services, that are not quite ready, but can help you move you further even as they are. If you cannot take risks, you will purchase only products and services, that have already proven their usefulness at other companies. Thus crippling the ability to innovate around you.

If the Australian government cared enough about this issue, they could do something about it - policy changes to encourage financial institutions to be less risk-averse, to encourage more of a domestic venture capital industry to grow. But it isn’t a political priority.

Australia consistently has R&D spending below the OECD average (as percent of GDP). The Australian government could easily fix this, either by increasing public sector R&D funding, or increasing tax concessions for private R&D funding, or both. But they don’t, because it isn’t a political priority.

(And doesn’t culture have a great role to play in determining what is seen as a political priority, and what isn’t?)


Do you have some concrete evidence for the claim that other countries’ failures are due to being "shut down by the US"?

Australia's failures in space aren't the fault of other countries, they are the fault of Australia. Australia's National Space Program (1986-1996) failed primarily due to chronic underfunding. That's not the fault of the US. And then Australia went over 20 years without any dedicated space agency – not the fault of the US either. And now Australia finally has one, it has a puny budget of approximately AU$10 million (around US$7 million) a year. If Australia was to spend as much on the Australian Space Agency (as a percentage of GDP) as the US spends on NASA, its budget would be over $1 billion a year. But it's not. Is that the fault of the US? If Australia decided to spend $1 billion a year on its space agency, I think NASA (and the US government more broadly) would be quite pleased. Australia could make some substantial contributions to one or more NASA programs.

In the 1970s, Australia was an associate member of ELDO, one of the predecessor agencies of the European Space Agency. The ESA was happy to offer Australia a similar membership status in ESA, but Australia declined. Canada went ahead and remains a formal “cooperating state” of ESA to this day. The offer is still open - the ESA has said so publicly [1] - but Australia isn't interested in taking up the same status as Canada, primarily because it would require Australia to contribute financially to ESA projects (i.e. spend money on space, something Australia doesn't want to do). (I've heard the claim that the US pressured Australia not to become an associate ESA member, but I've never heard any actual evidence to support that claim; and did they pressure Canada not to have their current “cooperating” status?)

[1] https://thewest.com.au/news/australia/europeans-offer-space-...

> Do you have some concrete evidence for the claim that other countries’ failures are due to being "shut down by the US"?

I'm actually surprised you're asking this, because this phenomenon is fairly well known, but anyway. TSR2, Black Arrow and Concorde (with France of course) are a few examples from the UK, and the Avro Arrow is an example from Canada. It does require capitulation from weak or compromised leaders in the producing nation. Take France as a counter example of a country that refused to capitulate to US demands e.g. military aircraft, rockets.

Again, you haven't actually responded with any hard evidence that these projects were "shut down by the US" – i.e. what exactly did the US do? when exactly did it do it? – just the claim that they were and that the (alleged) fact that they were is "fairly well known" (by whom?).

Concorde was always pushing the envelope in terms of technology and economics. I think a big part of why it ultimately failed, was that improvements in the luxuriousness of business class and first class made it more attractive for many people to spend less money on getting there slower in luxury than spending more to get there quicker but in Concorde's far less luxurious conditions. Add to that the small addressable market (due to the high price point and limited number of aircraft), the post-9/11 aviation downturn, the Air France Flight 4590 accident, the very high maintenance costs, the fact that it was only ever used by two airlines (whose Concorde operations were only profitable because the British and French governments had written off the lion's share of development costs), and Concorde's ageing technology. (The British and French governments were not willing to pay for Concorde 2.0, and the airlines could never afford to pay for the 1.0 version themselves anyway.)

In the case of TSR-2, the UK government cancelled it because they believed the US F-111 was cheaper. They chose to prioritise cost over the interests of their domestic aircraft industry. That may have been a foolishly shortsighted decision, but ultimately fault for it lies with the UK not the US.

> It does require capitulation from weak or compromised leaders in the producing nation. Take France as a counter example of a country that refused to capitulate

My original point was about how differences in culture can potentially lead to differences in economic outcomes. I think differences in culture could also play a role in whether a country's leaders are "weak".

> Innovation happens on the factory floor.

This! And more generally, innovation is more likely to come from people who are having to deal with messy reality. (See, e.g., the protocols for treating COVID-19 that have been figured out in emergency departments and ICUs all over the world.)

Not only that, but the US political system moving towards a much stronger anti-immigrant stance is literally reversing the brain drain and kicking out the best part of that deal, the smart people.

Eh, I dunno about the whole 'political system'. Maybe conservative populists, one of which just so happens to be president for now. After inauguration day in 2021, I expect he and everything he stood for will be discarded to the trash bin of history.

The US has had bouts of anti-immigration since it’s founding.


Yes. The White House has made this a very explicit part of their agenda since day one.

I think you might be confusing legal and illegal (or documented/undocumented if you think “illegal” is too on the nose) immigration.

You can be anti-illegal-immigration (which is effectively being anti open borders) without being a xenophobe that is wholly “anti-immigration”.

He has also reduced legal immigration options. Refugee caps much lower and the White House wants to get rid of the visa lottery. https://www.cotton.senate.gov/?p=press_release&id=603

They worked Cotton on this bill which would reduce legal immigration.

H1B Covid policy is anti-immigration.

Which country on earth has a COVID-19 policy that is not (to some extent) anti-immigration?

Yes, you're very unaware. Anti-immigration has been a signifigant part of the current administration's agenda (and longer term than that, politics in general on both sides of the aisle). I won't venture into the exact motivations for this stance (it's not helpful to the current discussion, and HN isn't really the best place for that kind of political talk) but the sheer pervasiveness and amount of vitriol behind it are the reason you're getting downvoted - it's hard to believe anyone legitimately hasn't seen it.

Here's a quick selection of related articles. Apologies that some of these may be paywalled. Also this was just from random googling, so I haven't fully vetted these links for accuracy/bias.




(2018) https://www.theguardian.com/us-news/2018/dec/21/us-immigrati...

(2017) https://www.migrationpolicy.org/article/top-10-2017-issue-1-...

(edited for formatting)

I've seen anti-illegal-immigration sentiment, absolutely. But what was said was "anti-immigration", which is not at all the same thing.

Even if you think US border policy should be less strict/militarized, that isn't relevant to this discussion, as the number of undocumented migrants working in state-of-the-art chip design is approximately zero, and that's not because of their legal status, it is because skilled labor (which is obviously what is required for chip design/fab) is very capable of getting a visa to live in the US under the current (and previous) administrations.

In fact, the current administration has said repeatedly and explicitly that they do not want to curb legal immigration, only undocumented/illegal migration.


A cursory reading of the headline might suggest this was to prevent spreading of the virus, but the president explicitly said that it was to protect american workers from immigrants competing for jobs

There was also the case a few weeks ago where the government told students on visas that they need to leave the country if their school switches to all online courses in light of the virus[1] which is not only anti immigrant but explicitly reversing the brain drain as was mentioned earlier in the thread. They did reverse that decision but only after major colleges started a joining in a major lawsuit.

There was also the whole "Muslim ban" limiting legal migration.

Oh and ICE deported an Iranian student despite being given a court order not to, so they're also explicitly breaking the law in order to deport people[2]

In both words and actions this administration has been against legal immigration even if it's at less of a degree than illegal immigration.



Fluctuations in handling of legal immigration in light of COVID-19 are literally present in every country on earth.

Referencing actions taken in these unprecedented circumstances as if they are the norm is incredibly disingenuous.

2 of the 4 situations were not Covid related. For the visa one, they had already lifted the restriction on visa students being able to attend online only courses and then rescinded that relief for no given reason, and backed down as soon as challenged.

They continually snipe at legal immigration and make it harder for people to legally immigrate when they get the chance. A claim was made that they were pro legal immigration. Here's the counter proof

> Innovation happens on the factory floor.

I don't particularly miss the day of colonialism, however I just don't see innovation happening on the floor at a chip fab.

It's got a stronger chance of occurring in the chip fab than in the CFO's office.

Practical experience matters even with high tech processes. People who live out in the abstract design space miss small but important details.

There are many more layers of engineering in a modern CPU than designing the ISA, the hardware layout, and even the initial mask. Core designs can be licensed. The process is a big part of what makes a processor perform well.

People engineer the cleanroom, the litho, the straining, the insulator layer, all sorts of things individually. Every part of a fab can be reengineered towards making a better, more consistent processor with better yields.

Are you implying these are local people making these adjustments at the fab?

Are you aware what sort of businesses Global Foundries, TSMC, and even the fabs of Samsung and TI are? Every part of the process has its own engineers. Some of them design the chip. Some of them design the fab. Some design the equipment in the fab. Some tweak the settings. All are willing, or should be, to take feedback on possible improvements and do small test runs between large batches. Anything that raises yields is a welcome improvement if it works.

> The U.S. thesis for outsourcing is that the manufacturing countries are filled with braindead automatons

So Wall Street was serving here as a template too?

But jest aside, if success isn't measured by long time goals we should have expected these results.

Still, as a trained engineer in the US+EU I went for a non engineering career out of university in the early 2000s as even here engineers are badly paid.

Out of curiosity what career path did you choose? And where is here if I may ask?

Finance in the UK

How can it be that engineers are poorly paid in the UK? From some googling it seems like the average is £62,500. That tells me the upper tier is probably around £80-100k, which is not too bad. Lower than Silicon Valley salaries, but surely living costs in the UK are also significantly lower if you live a bit outside of London?

£62k is not very far from entry level in finance, and as you become more senior you would typically earn 2-3x that. Outside of software engineering, I don’t think things have changed that much.

Part of me thinks that if engineering was offering better careers, we would have flying cars now. But I also see it as a market and if engineers aren’t paid that much, it is probably because there isn’t a supply shortage. For software developers, there is a supply shortage (as every company is becoming a software company) and it translates into higher wages.

"Part of me thinks that if engineering was offering better careers, we would have flying cars now."

> I was thinking something along those lines, we've had an entire generation of smart people go into finance and create financial innovation that led to 2008 (amongst other, potentially positive, things).

Glassdoor gives the following salary numbers for Google: London: £65000 (=$83.7k) Mountain View (CA): $145000

Both seem a little low; I'd expect a mid-level in London to be >£75k (~$100k). Less sure on the US salary; maybe >$155k? Neither include equity, but I imagine they scale pretty similarly.

So a London dev at Google is looking at a loss of $70k per year compared to their SV colleague.

London isn't cheap. It may not be NYC or Bay Area levels, but you can't really escape it by commuting in. You're going to be competing with everybody else trying to do that and knocking £5-6k off your post-tax income, so I don't think it really comes out as "cheaper" to try that. I was estimating needing ~£1000 per month as basic living costs (rent + public transport, but not food) in shared accommodation when I was looking to move there (in the end I didn't, but not because of anything I've said here).

Edit: Fix bad maths in comparison

I think your looking at software engineering roles, not traditional engineering roles.

Traditional engineer roles pay far less, think £40k in London.

As an American that just immigrated and graduated Uni this is really disheartening. I was planning on going into SWE in London. Is finance the better option? The market seems to be hot for developers.

Long term career in mind, finance is not necessarily the better. In the UK it's the last post-colonial metropol advantage it has, but it also will gradually evaporate due to Brexit, anti-corruption legislation and eventual introduction of rule of law in the 3rd world.

The City has one of, if not the world's greatest concentration of finance brainpower and capital, and has survived through hundreds of years of global turmoil as one of the world's leading financial centres. Do you really think these vested interests will just give up that lead and move elsewhere?

It became global leading when it happened to be the navel of the nascent Empire. With reducing global relevance of the United Kingdom and improvements in conditions for finance elsewhere there are fewer reasons for capital flows to be managed there.

Still may take a while and one may well live a fulfilling career in meantime, just saying it's far from a sure bet. Capital is not terribly sentimental.

No skin off my nose either way - I'd be quite happy to see house prices fall, which is one of the likely consequences.

There was a great concentration of brainpower and capital from the industrial revolution, from the age of sail, and from the victorian era.

That hasn't lead to the UK being a leader in railway construction or shipbuilding - despite the vested interests.

The unions bear a large part of the blame for destroying British industry.

Relying on what happened in the past as a metric for the future might be the slippery slope fallacy..

Yes, and the death of the UK and London has been predicted over and over, and yet somehow they continue to flourish...

Finance in London does pay well, but there are plenty of engineering opportunities as well.

I would say that UK junior and mid engineer roles are in general worse-paid than in the USA, but senior roles are competitive. Getting a big salary straight out of uni is not something I've seen over here but you hear about it in California.

If you're in London you have the benefit of a pretty thriving tech scene. If you want to look further afield then Bristol, Manchester, and Edinburgh seem to have a lot going on as well, and lower living costs.

Top tier tech is competitive with finance in London (FB/goog/DM/palantir/apple) and probably less work than equivalent technical finance roles (Jane Street, 2sigma, GR etc). But things if you're in the top 20% of your cohort in finance the money gets sillier much faster as I understand it. Non-technical finance roles seem to pay barely minimum wage when you divide by the hours for the first few years.

Yeah that was a decision made 20 years ago. I think developers can make similar salaries as people in finance (outside of a few hedge funds), particularly if you do contracting. Plus after the financial crisis, finance is a smaller club. But for other areas of engineering, I don’t think I would take a different decision now.

You can go into software engineering within finance.

> Pretty soon the countries we outsourced to will come up with better designs too.

Which, to me, would be a very good thing, so I wish them the best of luck. Many large companies that exist because of an earlier lead (established decades ago) and high barriers to entry would benefit from hard competition. My 2c.

> but now other nations are in a position to pay their engineers more than U.S. companies.

No. Please write 1 country.

You have to include real cost of living in this. When it comes to software engineering many people from Eastern Europe and I guess even India (Vietnam too) are relatively speaking better off compared to many of their peers in SV.

I live in Romania and I have a close friend who works locally for a big US software company that everyone on this website has heard of and he's about to purchase an apartment in one of the poshest areas of Bucharest (think one block away from Central Park, NYC), the money for doing that having been earned from that US company. No way he'd be able to do that in SV or in NYC itself were he to be employed directly in the States (unless he was a star engineer or something).

He won't probably be able to buy himself a Porsche or any other 100,000+ euros car for that matter, but those are mostly one-off gimmicks that pale when you compare them with the cost of living + healthcare (even in Romania, if you have decent money you can now get pretty decent healthcare).

I've heard that "the Chinese FANG" is willing to pay similar money to real FAANG for talent. Can anyone confirm/deny?

I live in Shanghai. I will eat a live chicken if there are companies paying Bay Area salaries to software engineers, Baidu, Ali Baba and Tencent included in significant numbers. Zoom has its programmers in China because they can get the same value for 30% the cost. Even entry level finance doesn’t pay as well here as in the US. At the top it does because it’s finance, you eat what you kill but there are to my knowledge no markets equivalent to the US in software anywhere. That’s why people keep moving. It’s why Lambda School abandoned the EU and international expansion. There’s much, much more money in the US.

What's the cost of living in Shanghai vs the Bay Area?

I know people who live on a couple hundred bucks a month in Shanghai and I've had satisfying meals for $2 there. I think if companies were paying anywhere near half of what Apple, etc paid in California, people driven solely by money (emphasis here) would get by better in Shanghai.

Right now, America's draw is living in America. People will take cuts to their actual earnings to live somewhere desirable. Whether America can hold onto that appeal in the coming decades is the big question. Money might not be enough to draw people in.

Depending on what ranking you use the USA is 7th, 8th or 9th in GDP per capita. All the countries that are richer are somewhere in size between a US state and city, several of them are petro stateside city states and two, Ireland and Luxembourg, have grossly inflated GDP by virtue of having so many multinational subsidiaries there, who pay very little in tax and contribute little to GNP.


This understates how rich the US is. By household final consumption the only place richer is Hong Kong and the US has substantially smaller households. It’s an almost $10K drop to third place Switzerland.


What draws people to the US isn’t a small difference in money, it’s a lot of money. $100,000 a year would be great money for an individual contributor here, $200,000 for a new grad is where Google starts in the SFBA. There is a cost of living difference but it just doesn’t come anywhere close to wiping out the huge difference in compensation.

There's something odd about this.

The US is said to have very high salaries in PPP terms. But at the same time, I read about extreme costs every so often. "3.2 km of tunnel […cost…] $4.45 billion. A 1.5-mile (2.4 km), $6 billion second phase […] is in planning". And about things that don't seem to coexist well with affluence, like large parts of the population being unable to go without a single month's income, or car loans at more than 100% of the purchase price and with repayment plans longer than the car's average lifetime.

Something won't fit tigether here, mentally.

In an environment where people have more money everything gets more expensive, but labor does it more and faster. That’s Baumol’s cost disease. Only very expensive internationally traded capital goods and financial products have a single global price. Haircuts and coffee are more expensive in the US and Switzerland than in Ethiopia. The US does have unusually high construction costs, but that’s incompetence, see Alon Levy’s work. But food, clothes and many other things are cheaper than in other developed countries and the US is a technological leader in more or less everything, as in among the top 3 countries.

People who can’t save and borrow on bad terms exist everywhere. The US is different from Germany, but those people exist there too. The US is not different enough for a special explanation to be needed.

It's not really so much incompetence on the developers' part. The US has an overlapping patchwork of jurisdictions and regulatory authority at the federal, state, county/parish, city, and often other levels (township, flood control district, sometimes a converged health department...). The number of permits, regulations, inspections, and taxes on top of any project often costs as much as the materials and sometimes as much as the labor.

Isn't all that a fancy way of saying that comparing income using PPP rather than exchange rates makes sense?

Is there a list of countries by median household final consumption expenditure per capita after debt and medical expenses ?

Median, no, but average household consumption already includes all consumption, market and non market. The average USAn with negative net worth and no job lives a lot better and consumes more resources than the average Sudanese person. The average employed USAn lives better than probably the top 2% of Indians, definitely the top 10%.

Define living.

I think the only developer market close to the US here is the financial sector in London. If you have solid experience in a desirable skill and finance experience you can get really good money. Outside that the pay is decent and I'm sure there are small pockets of specialists in small niches doing really well, but there's no FAANG level market.

> I think the only developer market close to the US here is the financial sector in London.

Depending on which "US" you mean. If you mean regular US senior developer salaries (around $150k) then they're achievable, on a contract basis, in pretty much all major centers in Europe (London, Paris, Munich, Zurich, Amsterdam etc.). If you mean the $500k FAANG salaries, then I've yet to see an equivalent in Europe.

The “on a contract basis” is important. They’re not available as an employee. Contractors in the US get paid even more. Compare like with like.

Does it matter, especially now? It's not like engineers can move freely to the US

If you're getting a "millionaire" salary in India or China does it matter that's not equal to a US salary in dollars?

Yes it does. A good way to getting "millionaire" salary in the world is to work remotely for US/UK companies. In the "new" remote age, it matters.

But working remotely from another country, it's much more likely that the engineer will work for a local company (or another international competitor) after that. Engineers that move to SV for a job are likely to work for US companies for a long time.

For the individual company the difference will be small, for the US as a whole it's a big difference.

Switzerland? Aiui (on 2nd hand knowledge) salary for a SWE ranges from 80k usd to 250k usd

The top of your range would be the bottom of the range for a Google beginner with 3-5 years of experience.

But Google is in Switzerland too. It probably doesn't make sense to compare country level averages to the pay off a single company famous for huge comp.

The question was whether you can get FAANG like compensation in other countries. While Google has an office in Switzerland, it is minuscule in comparison to the US.

There are over 2000 workers there. That's "miniscule" compared to the HQ in California of course, but it's the biggest engineering office they have in Europe. The city it's in only has about 350,000 people, so it's a non-trivial local employer.

But we're not comparing country averages, we're comparing the highly paid spots only. The same multinational company e.g. Google will pay you much less in Switzerland than they would in California.

I don't know where you're getting your info from, but Google salaries in Switzerland are extremely competitive with California. The average for software engineers at Google Zürich is around $300K at L5 I'd say. And the housing situation, while still very expensive, is a lot less fucked up than San Francisco.

Source: I'm a Googler in Zürich. Current total comp is at $320K after 3 years at L5.

This is insane

You can get that in a bunch of places outside the US (Sydney, London at least).

Don't compare gross income, but net income (after tax, healthcare and especially ridiculous COL) plus the recent anti-immigration swing and chaos in politics - and suddenly the US are not as competitive for foreign talent any more, and Japan and Taiwan certainly have the money to compete.

I only compare gross income because I work remotely. My healthcare costs are the same if I work for German client or US client. And German pay sucks.

To be fair, that's also because with US salaries, you get a rate that's adjusted to a much higher cost of living. If you adjust for that, you can get similar salaries as a programmer in finance or pharma in Germany.

I completely agree that overall salaries are lower in Germany even adjusted for cost of living but the insane price levels in SV compared to most parts of Germany skew that even more.

The situation is more severe in the UK where central London has similar cost of living as NYC or SV but salaries at a quarter of SV levels outside of finance.

I have ignored remote work for exactly this reason - it's obvious that someone on a six figure SV salary can live like a king in Asia but in the SV still may be forced to live in a car (both literally as in rough sleeping or figuratively as in being stuck in commute hours a day). But jobs and people capable and willing to do this are rare in absolute numbers.

I don't have a lot of experience and confidence regarding this, but pretty sure DE,NL,ES,IT pay sucks compared to US as an example (as a local).

You can't put German and Dutch salaries in the same categories as Spain and Italy. There's easily a 50% difference.

German salaries are really attractive for Europe.

It still sucks compared to US.

Even after taking out cost of rent, cost of living and cost of healthcare? I highly doubt that.

Having worked in both places, yes, they still suck in comparison.

In absolute numbers German pay may suck, but once you factor in lower rents (e.g. in Munich a 60 m2 2-br flat in the city center will set you back ~1200 €), lower healthcare costs (no co pays aside of 5€/Rx, no "out of pocket" costs for medically indicated procedures), a decent pension scheme and lower cost of food and other expenses, suddenly it is competitive.

I'm also in Germany for the moment. How did you get US remote jobs if I might ask?

You just apply ? There are a lot of job boards & places to find remote jobs. I only work in CET and not all of them allow this. Company I work now for example has developers in different timezones. You will have a meeting once a week at 19-20 though.

UK is also much better and the timezone is ~same. I've mostly worked with US, UK, CA, IR (subsidiary of US).

An easier way is to focus your skills and specialize in, say, scalability/search-engine/low-level/ml. There will be less jobs, (compared to js), but you will have more leverage.

Switzerland comes to mind. They might not compete with FAANG comp, but the average engineer is better paid there than the average engineer in the US. I'll admit, they are an anomaly.

> Maybe repeat by having your suppliers do more and more of the engineering. At a certain point, you are just a brand, with no actual engineering capability.

Don't they teach this in business school? It just seems incredibly obvious to anyone with an engineering background.

They do actually teach this in business school, but not in the way you're thinking. There's a strong business option* where marketing (and thus brand) is more important than actual engineering or innovation. So under this strategy, if you are just a "premium" brand with a mostly outsourced and relatively cheap product line, that's a good thing because your margins will be high. You can actually see this happen to a lot of brands over time that were once trusted for quality and are now kind of garbage (ex. Craftsman tools). It is, of course, a short-term strategy because eventually the market figures it out and the party ends, but while it works shareholders (and executives) are quite happy.

*Note that business school teaches other options, strategies, tools, etc. as well, it's not like everyone is trained from day one to pump and dump.

Source: went to business school.

Yes I've seen it happen many times with products. So it's a conscious decision by the CEO, Board and execs to extract maximum value for themselves and shareholders and do a runner? I imagine they couch it in different terms in the minutes.

Edit: I probably didn't phrase that well - how would this happen, is it a nudge and a wink thing to the board, or the board is clueless and the CEO / execs just decide etc?

It's more of a strategy to maximise short-term gains, and the downsides aren't considered - because they don't need to be.

Public shareholders have absolutely no loyalty to any company they extract wealth from, and they can move their capital elsewhere in seconds. And CxO/execs are typically awarded bonuses on share price movements.

So any action that moves the needle on the share price is a good thing. If this has bad long-term consequences, CxO/execs can always try to bail if they see incoming. They don't always succeed, but they have a fair chance. Employees are more likely to be left without lifeboats in the sinking ship.

This still seems like foolish thing to do long term.

What happens if we move all the profits and manufacturing capacity to markets Western investors are locked out of?

Who cares? I don't mean it in a snarky way, more so I am curious as which groups of people actually get affected. Workers? Certainly. Capitalists? Not so sure. Somebody in charge is making these decisions and they deem them to be good decision, otherwise they wouldn't make them

The shareholders need to be put in the dark otherwise the stock would drop since there is a explicit short term pump and dump scheme. Surely it has to be at executive and board level to maximize bonuses not in the interest of the stock holders.

Sound like most other shorttermist problems we are suffering, i.e. climate change, debt, etc.

I think your terms are spot on.

John Sterman used to say something very similar going back to at least the 2000 and the trend has only accelerated since then.

At the time I remember him giving lectures on System Dynamics models to demonstrate the effects.

Thanks, I'll have a look at his works

Good insight, I feel like engineers/developers would benefit from basic business classes, just to understand the thinking that can lead to the downfall of great engineering companies and hopefully how to avoid it.

Having recently seen this first hand, i.e. great developers being shuffled around and questionable business decisions carried out without a strong engineering buy in, it sucks.

> the downfall of great engineering companies and hopefully how to avoid it.

why is this the responsibility of the engineer (presumably, with no real equity in the company)?

The ability to choose the ship you set sail on and to know when it starts sinking.

My understanding on how to avoid it would be, if you see decisions from management that point in that direction, run. Or raise concern.

> You can actually see this happen to a lot of brands over time that were once trusted for quality and are now kind of garbage

You can see this in Silicon Valley as well with the number of startups that are absolutely worthless but have exorbitant valuations because because they came up with a metric that everyone fell for.

It's the tyranny of the spreadsheet.

The thinking isn't that long-term, it's focused on the next 2-6 quarters. Plenty of people got rich off of plundering Hewlett-Packard; to the point there's no meat left on that carcass aside from garbage printers.

Ironically enough, the test equipment division (what I would call the actual successors of the company) have been through 2 restructurings (HP->Agilent->Keysight). Almost as though having cutting-edge engineering & R&D will have the side-effect of making good products. Shame that costs money though.

I'm not sure I'd describe Keysight as a success story of cutting edge R&D, I can't speak for the current Agilent as I'm not a chemist. They and other namesake test equipment brands seem to only focus on the ultra high end and the lower to even midrange models are frequently white labeled products from overseas competitors with a more polished UI. And they're not even always pushing their top end stuff further, their top end multimeter is virtually unchanged since 1990.

Oh, Keysight gear doesn't hold a candle to vintage Hewlett-Packard. However, I just wanted to argue that Keysight is closer to Bill & Dave's garage than the current HP incarnation. They have some OK gear, but too much of it seems overpriced; they are definitely letting the cheaper Chinese manufacturers catch up to them and seem to only want to rely on institutional contracts.

I personally don't know anyone in the hobbyist electronics space that owns any Keysight gear, my only exposure has been in university labs. Bit of a shame, really.

Yeah it's kind of crazy what HP was able to accomplish, the reason keysight hasn't updated a lot is because they didn't really have to. The DAQs are incremental improvements from what I've seen, the aforementioned 3458A still competes with new instruments, and I honestly enjoyed the 54601A scopes I used in school more than the new Tek ones I have at work.

They are not as affordable as chinese counterparts (Rigol or Siglent) but their 1000x line of scopes is a step in a good direction. I own one although I paid more for less features as compared to Rigol, difference in quality and polish of the UI is staggering.

The 3458A is still one of the highest accuracy, most stable and linear multimeters around, even though new 8.5 digit DMMs have been introduced since then. Anything significantly better is going to look more like a physics experiment than a neatly packaged bench instrument.

I'd appreciate something other than a GPIB connection in 2020, but you're right it's really a marvel of a design. Since getting it in the lab I've started reading more about how it works and there's a lot to learn. The widely lauded ADC in particular is really impressive.


>Plenty of people got rich off of plundering Hewlett-Packard;

certainly not the employees!

C-suite's job is, unfortunately, only to get the shareholders rich, not the employees.

maybe thats part of the problem, I recall a trend where employees bought out the company and ran it. Not sure what happened to it.

We can't let the great unwashed own the company and run it for God's sake. /s

That's what happened.

There have been a only few instances of employee ownership of large corporations, never any sustained trend. United Airlines employees took over majority ownership in 1994 but that never really worked and they ended up bankrupt.

I suppose HP were challenged externally and couldn't adapt, the money savers start moving in, which pumps it up temporarily and it becomes a vicious cycle.

More obvious in retrospect and with experience, perhaps.

In the 1980s or so gadfly (relative to IEEE) Irwin Feerst was sounding a fairly lone warning for US engineers: Outsource packaging, then you will outsource fab. Outsource fab, then you will outsource process engineering. Outsource those, then you will outsource design. ...


Almost all large, modern, public businesses are engaged in a pump and dump scheme where no one making decisions has any intention of being around for these last phases of corporate decay and death. It seems nuts to us from the outside and is super damaging to the long-term interests we all have in the US but this is taught in business school and is in fact the right business decision for these people to make given the context and sociopathic motivations they operate under.

No US business school teaches pump and dump schemes. The fault lies with corporate boards who set up compensation plans not aligned with the interests of long-term shareholders. It's the basic principal / agent problem.

Ok ok schools don't literally teach that, true, I was ranting but "corporate boards who set up compensation plans not aligned with the interests of long-term shareholders", is a pump and dump, exactly my main point, which stands.

Unless you're vertically integrated, manufacturing is incredibly difficult to make money from. The traditional South Asian/East Asian medium sized manufacturer is run by people who put in outsized efforts to the returns. The margins are razor thin in most units, and the competition cut-throat. The units simply won't make money if they're delegated to MBA types, because the true cost of managing the unit is much much higher.

It's actually the perfect small/medium business endeavor in developing countries because of the labor cost arbitrage and economic opportunity. But a typical college educated American entrepreneur will not find it appealing, compared to starting a tech business or running a services business.

Might be controversial to say, but why manufacturing cost is so high in US they making outsourcing so profitable?

A Taiwanese silicon engineer salary, even the top band is probably only higher than the entry level of a web developer in SF.

Something is definitely wrong

> Might be controversial to say, but why manufacturing cost is so high in US they making outsourcing so profitable?

I suspect that the falling rate of interest is the primary cause.

The falling interest rate on house loans makes houses increasingly expensive, as potential home owners can afford more expensive houses (given the same income) the less they have to pay in interest. The same principle ought to apply to companies who buy factories for manufacturing, with the primary difference being that housing cannot be outsourced while manufacturing can.

This hypothesis could be substantiated by looking at the degree to which countries with a long term falling interest rate outsource to countries with a long term stable or rising interest rate, and vice versa.

High cost of labor as it's a rich country with an expensive currency, unions, and in some places, protective labor laws.

I think you're getting downvoted because you're not allowed to imply unions and protective labor laws drive up the cost of labor in this country.

There are certainly a lot of people on this site who seem rather blind to the issue.

Foxconn's million+ workers do theoretically have a union, but only since 2007, and with actually elected leaders only since 2013. At that time they still faced the issue that most workers didn't know they were meant to elect a union representative at all and the union committee leaders were all managers. What little workers rights exist come mostly due to pressure from Apple and American unions.

And of course, as China is getting richer and the workers are starting to slowly come to resemble western workforces, companies are relocating to Vietnam.

Web developers in SF are barely representative of web developers outside of SF, let alone factory workers in the US. That's a bad comparison.

It is not IMO.

Web developer in US can work freely in HoL area if they want to relocate.

This is how we lost a lot of redundancy in the rail system too. Why maintain three routes or two tracks per route when you can boost your gains for a small chance of going out of business?

Reminds me of the old joke about the ideal company is having one person to turn on a factory and then place a logo on it, while engineering and everything else is outsourced.

It's like celebrities launching lines of cosmetics products - all they do is contribute a picture of their face, and then they rake in the money.

This is how software works, too, in some cases even open source, look at the Python ecosystem. Search for the actual authors or who is doing the work vs. who is going on stage in the Hyatt.

Sadly a very effective strategy until you develop critical thinking, some struggle their entire life to achieve it.

Critical thinking can suggest that the actual physical product lacks innovation or doesn't bring anything dramatically better on table but the branding does, therefore the branding people take the money.

Not exactly, they also signal that the style of the product will overall match the public image of the celebrity, which may actually save some time whey you are shopping, but most probably not worth it.

On the other hand those employees are a lot easier to fire at extremely short notice and virtually no cost than any other developed country, so there's that.

The corporate tax rate was lowered from 35% to 21% by the Trump administration to make it in line with the EU. (Though some EU nations have a corporate tax rates as low as 9-15%)

Do companies have any excuse now regarding using taxation as an excuse to move manufacturing outside the US ?

The issue was not so much the tax rate but the double taxation, iirc. Lower tax rates help bit isn't solving the underlying issue.

There is no such thing as "double taxation", because tax rates are not additive, thanks to tax treaties. So, if the tax rate is 30% in US and 20% in EU and you've already paid EU tax, you only owe IRS the difference (10%) not full US rate.

Since, thanks to Trump, US tax rate is now almost equal to European, then you pay almost no additional tax for bringing your profits back to US.

The only remaining issue is additional accounting required, meaning you have to hire accountants both in EU and in States.

When I hear about “double taxation” in the U.S. I think of the taxation of corporate profits combined with income taxes. For capital invested in stocks, I am taxed twice on any income derived.

Once at 0% in Ireland and once at 15% for capital gains. Where can I sign up to have my income double taxed like this?

My understanding is that this is only possible for very profitable corporations. Can smaller entities get away with this?

You are taxed only once: with income tax. Company is also taxed only once: with corporate profit tax. The reason why that makes sense is because corporation is treated a separate legal entity (person), and that's a good thing for you, because you are not responsible for company's mistakes (that's why they are called "limited liability companies". Publicly traded companies also have limited liability).

Also, if, as a majority shareholder you appoint yourself as the CEO and pay yourself a salary, it is tax deductible for your company, so money that goes to your salary is taxed only once.

That's a reasonable first thought.

When I hear "double taxation" in the US I usually think first of self-employed people sometimes having to pay both income and payroll tax compared to an otherwise employed person only paying income tax.

Then maybe I think about places like Missouri where after you take your post-tax income and buy a car with sales tax on the purchase, you pay yearly for that car as a personal property tax (which does not include your license plate, or inspection, or the city sticker some local places require).

That's true for most countries. For the USA those tax treaties for some reason didn't work like that and double taxation of corporate profits does occur.


Under its current system the United States taxes the worldwide income of its citizens, including corporations. Foreign income is taxed by the source country and then taxed again by the U.S. upon repatriation

Because there are credits against the double taxation, if US tax rates = the tax rate in that jurisdiction then it cancels out. But many places have different rates and then the double taxation kicks in again - they get taxed on their income once abroad, and again if they bring the money back to the USA, so they don't do it. Changing the rates is a band-aid: most countries don't do this kind of double taxation to begin with.

> most countries don't do this kind of double taxation to begin with.

They would if they could, but they don't have ICBMs or aircraft carrier fleet to support their claims.

> You can't compete with the people who actually know how to make the things. Then you die. See Hewlett Packard.

Not if you are Apple

Deindustrialization is great, most industrial jobs are the worst. To the extent other countries want to take them on, it puts America at a huge advantage. Once robotization and full automation happen, there will be no reason to build in one place over another, and all the assembly work will return (as not having to expend energy moving final products from their assembly location to their final destinations will represent an advantage).

The jobs never will though, nor do we actually want them to. This is why we need UBI.

Seriously, why would we want, in the future, to pay people to smack metal with a hammer? Let's get the jobs elsewhere ASAP, unless you can tell me with a straight face the future will involve more humans in the manufacture of goods -- in which case we should be getting a head start. Nobody will say that though because it's crazy.

Trying to add humans back to the manufacturing process, who don't need to be there, amounts to a regressive make-work project we're engaging in, intentionally adding inefficiency, instead of actually changing society and positioning ourselves for the inescapable future of humanity. In my opinion, America is squandering a huge lead by glamorizing crap jobs nobody actually needs to do, let alone wants. Like coal mining.

What's next, a return to whale oil?

You are making a very large assumption about what types of work people are willing to do and dangerously that losing these opportunities is predetermined as an advantage for the US. Making strong statements like that and then ridiculing any opposition in advance of a reply is antithetical to reasonable debate.

Automation takes time to develop and can leave you inflexible depending on the task even with the wide variety of programmable robotics. Some products may not have profit margins, expected sales or complexity high enough where automation is worth investing in. You don't want to be in a situation where you need 300,000 of something in 3 weeks and the only way to get them is from a country that isn't an ally or refuses to trade with you.

Humans can easily do some things immediately and machines can easily do other things immediately. The more universally applicable a machine is, the more expensive it may be and possibly even priced against the highest value tasks it can do.

Universal Basic Income has many problems that aren't solved and there are many potential negative side effects, so it's premature to promote it confidently.

> Automation takes time to develop […]

And capital! The human beings who spend their time developing new technology need to be fed, which requires consuming someone’s savings.

Whether automation makes sense always depends on its benefit versus the cost of development. E.g. if we spend money on feeding 100 engineers for 10 years while they develop a technology that increases the efficiency of producing something by X% is it worth it?

We always need to make this calculation because the amount of savings is finite, so we have an interest in spending it in a way that creates the most gain.

> The more universally applicable a machine is, the more expensive it may be and possibly even priced against the highest value tasks it can do.

It's even worse: The more universally applicable a machine is, the more expensive its outputs are, because it is less efficient. Have you seen an aluminum can lid stamping machine? [1] It can only produce one thing, but that thing is incredibly cheap on account of the machine producing tens of thousands of items per hour.

[1] That's probably not its official name...

You're ignoring the rest of the supply chain and localization of talent. If I run out of a chip in the states it might be a week or two of the line being down, in Shenzen I can have more that day. Similarly if I need to have a test fixture made or modified on short notice where do you think the talent is? Even if a US shop could do it better, which is a big if, can they do it faster? Cheaper? And why wouldn't I want engineers located near my assembly line close to the previously mentioned chips for the install? There's quite a few steps between design and final assembly, and for a lot of products assembly is much more than smacking something with a hammer.

EDIT- I'd add that this isn't just about China or moving manufacturing to the cheapest location. Let's say I need a specialty injection molded part with tough requirements for med device. The best shop might be in the US, or maybe Germany, or maybe Mexicali. Automation isn't going to change that.

3D print it.

Yes automation solves it.

Very few useful tools or machines are 3d-printable, due to tolerance and strength constraints.

Not trying to be rude here, but have you designed a plastic part for production? This isn't remotely practical for many if not most applications that I have seen. Cost, tolerance, surface finish, yield are all major issues. It's also far less automatable than injection molding. Post processing an SLA part for example is very manual, injection molded parts come out of the tool ready for use. What do you think is going to change about that?

It is not a universal solution, both because it's slow, and because of various limitations and faults it introduces. There are dozens of manufacturing processes, some of them fairly new and modern, that will never go away.

Why do you say that people don't want to work on Industrial jobs and they are the "worst"? I see this point of view regularly regurgitated on HN and I am genuinely puzzled.

That might be true for HN (folks who prefer working at a desk) but that is NOT true for a vast majority of people.

I have both friends and family who love to work on the factory floor/shop - it is more varied and interesting for them. More interaction, more visible output of labour and they don't need to join a Gym as they get enough physical movement in their daily job. They despise the idea of sitting at a desk the whole day.

A lack of visible output bothered me so much that I had to compensate with hobbies.

> That might be true for HN (folks who prefer working at a desk)

I think it's more so people who never worked a physical job and imagine those to be just the worst.

Except those robotic factories are going to spring up right next to, and in place of, the existing factories and will plug right into the distribution infrastructure that already exists. The path of least resistance is not to bootstrap a completely new manufacturing and distribution system in a completely different place (the US). No matter how tempting it seems from your efficiency point of view, things sprout from existing buds. America chopped off all its buds and is left languishing at the mercy of its suppliers. That was a huge strategic mistake.

Is that an attitude of an over-privileged, pampered SV person? Because it certainly has nothing to do with reality. Maybe you should go traveling for some time and meet people who live in the real world, not a bubble.

Not at all, I think those people should be paid, but I think their jobs should be automated. Whether it's true today, or in the near future, we can all agree: those jobs are better done by robots. Those people deserve to live happy lives, and so we should pay them out accordingly, but we don't need to introduce the inefficiency to do so.

Even Walmart is looking at replacing all their cashiers with robots. [1] Do you think assembly workers are immune for some reason? We need to skate to where the puck is going to be, not where its was 30 years ago.

I find the idea of my comment coming from a "pampered, privileged tech worker" perspective interesting, as I'm offering a part of my pay so that the folks doing the mundane, back-breaking work inefficiently, today, could be paid to stay home and look after their families or pursue self-actualization.

[1] https://bgr.com/2020/06/17/walmart-self-checkout-no-cashiers...

>Whether it's true today, or in the near future, we can all agree: those jobs are better done by robots.

It's already been done, except replace the "robot" meme with "industrial engineering" reality. instead of iRobot it's the machine that welds cars or makes pringles. It was done 30+ years ago. This world is distinct from the robot replacement headline clickbait we've been reading our whole lives.

>Even Walmart is looking at replacing all their cashiers with robots.

The technology for this has existed for 20+ years but it's always being "explored", that should tell you something. That's a distraction though. We're talking about chip fab, then you started talking about assembly line workers (which has zero meaning - can you think of a single assembly function that hasn't been automated 20 years ago?). Factory work today is skilled work, industrial engineering hasn't stood still.

>I'm offering a part of my pay so that the folks doing the mundane, back-breaking work inefficiently, today, could be paid to stay home and look after their families or pursue self-actualization.

you're getting ripped off. Modern factory jobs are pretty good. Self Actualization is a meme.

Your comment got downvoted because there's a lot truth in it. :D

Manual jobs will be getting automated, and countries relying on manufacturing will be hit by that process the hardest.

Like in Factorio - first you're resources bound, than you are labor bound, and finally you're design-bound.

Modern economy is design-bound. USA being de-industrialized developed economy offers good quality of life and opportunity for intellectual workers, and attracts a lot of talent. Most HN readers is Americans, so they love to complain about USA, but "grass is always greener on the other side". As much as there are messed up things in USA, there's also plenty of things that are working really well.

The main problem is the gap between people who benefit from the post-industrial economy, and those who do not. I also don't see a better way out of it than just some well-designed form of UBI. And considering all the COVID stimulus, the USG is already paving the way...

Yeah it'll be great once the countries that manufacture things starts producing multifunction robots that build their own supply chain and automatically revert to US control.

You don't need to do the assembly to design and take delivery of multifunction robots. Just ask ABB [1]. The IP, the design, and the schematics are the valuable part, the assembly is totally and utterly commodity. A 3D printer, a diffuser, etc, is equally effective in any country it happens to be located, and UPS is happy to meet you there for pickup.

We've started to see this return already, with Foxconn and TSMC setting up factories in the US (with no people in them of course -- because why would they put people in them?).

[1] https://new.abb.com/products/robotics

> Interestingly, it seems the US is almost unique in our mass deindustrialization

The US manufacturing share of real GDP has been pretty much unchanged since the 40s - ie it’s growth has kept up with the rest of the economy.


Wasn't this investigated and found to be a misleading statistic?

For example:


The graph in this article, from BEA data, shows that when you don't apply inflation tricks, manufacturing has dropped from 28% in the 50s to 12% today. And they're still factoring in adjustments due to the "increase in quality of computers", ie, one computer today is worth some large number of 1980s computers.

It depends on the context - yes the number of manufacturing jobs has decreased (important in one context) but the industry at the macroeconomic level has been growing (to about keep up with everything else)

Also - when did accounting for inflation become a ‘trick’? No way you can look at nominal gdp on these time frames.

Did you look at the article I linked? It's not really focused on the jobs aspect, on which the data is clear, but rather on the accounting, and in particular your graph which is always cited in these discussions. I would need to quote too long of an excerpt to make a full reply. Here's where the discussion starts though.

> Real gross domestic product is GDP adjusted for inflation, which allows us to better measure and compare growth over time. If you're measuring and comparing different industries' shares of GDP over time, it's not immediately obvious why you would need to do such an inflation adjustment. The very act of dividing manufacturing's contribution to GDP in 1960 by overall GDP in 1960 does the inflation adjustment for you, given that both amounts are in 1960 dollars.

> The BEA's remedy to the problem is to put up warnings against doing share-of-real-GDP calculations all over its website.

It's worth reading the whole thing.

Comparing manufacturing output between time periods doesn't make a lot sense. How many 1980s mainframes equals one iPhone? It's a woefully incomplete question if you goal is to reason about economic value/output. Computing is not particularly special in this respect...

The manufacturing/services divide itself is a bit silly without careful attention, and I think is incredibly overplayed because "manufacturing" aligns with a bunch of other political/economic geography stuff.

This is true everywhere, not just in computing where it's most obvious. Even agriculture in a lot of ways.

By the time you apply all that careful attention, any sort of meaningful macro work becomes impossible. You're just doing a whole ton of little sector-by-sector analyses.

And it gets worse if you start looking at the labor economics instead of just measuring output. It used to be that manufacturing-related services and the physical work of manufacturing were relatively coupled. The people working in the manufacturing plants more-or-less knew how all the equipment worked. That's not really at all the case anymore.

> Comparing manufacturing output between time periods doesn't make a lot sense. How many 1980s mainframes equals one iPhone? It's a woefully incomplete question if you goal is to reason about economic value/output.

I believe that was the point of the "share-of-real-GDP" adjustments. Because "prices of manufactured goods have gone up more slowly than those of other goods and services (think health care), Baily and Bosworth reasoned that manufacturing's shrinking share of nominal GDP understates its continuing impact on the economy."

Apparently there is a decent way, at least in principle, to compare mainframes to iPhones; that is, to compare their contributions to improved productivity independent of nominal prices. But it breaks down over longer time scales unless you can plugin real numbers for changes in relative value-add across entire industries, which in this case isn't possible because the trick was used to adjust industries in the aggregate by assuming a constant ratio of value-add.

Why have we immediately reduced the conversation to the "economic" value of manufacturing? There is also a strategic geopolitical advantage to retaining domestic manufacturing, which is the most relevant part of this discussion.

Jobs, the economy... there is more to life, and more to maintaining a hegemony.

> There is also a strategic geopolitical advantage to retaining domestic manufacturing, which is the most relevant part of this discussion.

NB: there's also a strategic geopolitical advantage to out-sourcing manufacturing. The US would be a weaker nation if we had to commit a bunch of human capital to textiles.

But to answer your actual question:

> Jobs, the economy... there is more to life, and more to maintaining a hegemony.

You could say the same about freedom. Most people would prefer a good-paying job over global hegemony, and care about the latter only in service to the former.

note: I agree there's a strategic national security/economic/sov. interest in domestic chip manufacturing, hegemony or not. But the general analysis that we should make ourselves poorer in order to be world hegemon is, to most Americans, I suspect a bit backward.

Will we be more wealthy if we lose the ability to create the most advanced chips? I think not.

It's not hegemony for the sake of hegemony, but hegemony for the benefits of hegemony, which in our recent history have included technological and economic supremacy, both tightly correlated to one another.

I think globalism can be a good thing, and I hardly oppose the notion of outsourcing some economic activities. However, the original point was regarding ceding our production capabilities for a short term and potentially insignificant improvement in the performance of a single company.

> The US would be a weaker nation if we had to commit a bunch of human capital to textiles.

There is something to be said for balance, taste, and sensibility. Like, it would be cool if we were able to manufacture essential medical PPE during a pandemic. Them be textiles, my friend.

If we are unprepared for a crisis that can gut our economy because we were prioritizing the economy over preparedness... I'm sorry, but that is stupid and short sighted.

Jobs are important. The economy is important. But it's as if all nuance is lost on the GDP-measuring mouth breathers. You want a strong economy? Maintain the hegemony. Be a global leader in many diverse categories.

> Will we be more wealthy if we lose the ability to create the most advanced chips? I think not.

The post you're relying to literally says the same thing...

> medical PPE be textiles, my friend.

1. No. "Textiles" is a term of art, and it doesn't include N95 masks or 3 ply surgical masks. Full stop. Might include some basic cloth masks that are equivalent to a ripped-off tshirt sleeve with some elastic bands. But we never had anything approaching a shortage of t-shirts to rip up for cloth face coverings... what we're missing is the stuff with melt-blown layers and/or filters.

2. We do make N95 masks and could have surged production... shortages of the most needed types of masks in the US were due to an utter failure of political leadership, not a lack of manufacturing capacity. E.g., https://www.washingtonpost.com/investigations/in-the-early-d...

3. The same is true for 3-ply surgical masks. Repurposing a huge number of t-shirt factories would've taken ages and $$$. The difficult part is the melt-blown layer, which is more like N95 or surgical gowns than t-shirts. Not even clear having an existing t-shirt factory would even speed things up much, since, obviously, the barrier isn't "stitching a couple pieces of cloth together". But even if it did, what are the odds on that happening if we couldn't even give a green light on flipping the switch on existing N95 lines in May? Again, leadership problem, not manufacturing capacity problem.

4. All of that aside, this is the sort of thing you solve with strategic stockpiles and planning, not "make everything ourselves and hope to god some of the other lines can be repurposed ASAP" and especially not "constantly maintain manufacturing capacity and raw resource supply chains for every conceivable long-tail event". No. You figure out for each long tail event how long it takes to ramp up production, and then you stockpile for that amount of time.

The rest of the post is sort of _exactly_ demonstrative of the point I was making in my original post about macro analyses being limited and needing to go sector by sector. So I'm really not quite sure what we're arguing about here.

> The US would be a weaker nation if we had to commit a bunch of human capital to textiles.

While I probably agree with the broader idea, I can't help but think that at root, sentiments such as these are the root cause of our industrialization and significant but hand-waved human costs here that have been incurred.

The problem here is these GDP measures measure value add. Let's work through an example:

Scenario A: You assemble everything locally with 50% of your components from east asian component manufacturers, the rest from the domestic market. Your costs are $800 per unit. Then you spend $200 on in-house design and sell your product for $1500, for a profit of $500 per unit and a domestic value add of $700 per unit. At that price point, you sell 2/3 of a million units, adding roughly half a billion to domestic manufacturing.

Scenario B: You outsource all of your assembly to China and parts from East Asian component manufacturers and only do design/QA in-house. If you pay $300 to your suppliers, spend $200 on in-house design but sell your product for $1000, you have $700 in domestic value add. Say at $1000 you sell a million units, so you've contributed a billion dollars to the manufacturing share of GDP.

Now if a company switches from Scenario A to Scenario B, most people would (rightly, I believe) assume that some manufacturing has left the economy. But actually the manufacturing share of GDP has increased! However it's a very fragile increase, because a change in the terms of trade and suddenly you pay more for your foreign inputs and at that point it will appear as if manufacturing's share of GDP has fallen. There is also the issue that foreign rivals are going to be in a better position to compete against you, so long term this does not bode well for manufacturing's share of GDP.

Now this is just one company, and buying from domestic suppliers also contributes to GDP. But they have the same issue -- they can outsource production leaving design/QA in house as well. So you can have a situation with high manufacturing share of GDP with nothing produced in the nation other than design/QA. So all these GDP by industry measures, at the end of the day, are measuring industry margins more than actual manufacturing per se, and really weren't designed with outsourcing in mind, and aren't intended to be used in a time series analysis during which a lot of outsourcing is happening.

Interesting. Are there better ways of measuring the contribution of manufacturing to the GDP? Or is it simply not possible by definition?

I'm not in a position to say that no better way is possible, but all ways of defining economic aggregates run up against conceptual problems when you try to apply them to an actual economy, the most intractable of which is how to assign a single number to a changing mix of heterogeneous goods. The difficulties with defining "inflation" when the basket of consumer goods changes every period is well known, but you have similar difficulties with defining production if you change your inputs -- of which outsourcing is a special case.

In most situations you end up using price as an aggregator, even if the price is itself normalized by a scaling factor. That means if you shut down a factory today and have your goods produced cheaply overseas except for a few small finishing steps here, it will be the relative prices of the inputs that determine how much of the good was produced here versus how much was produced overseas, and this wont generally reflect the "real" amount produced here versus overseas as viewed by a layman. In fact, if a producer couldn't outsource cheaply, they probably wouldn't do it, so there will be a consistent gap in terms of what the laymen sees and what the BEA data will show. And note this also creates the appearance of a productivity boost. Imagine Apple firing all their manufacturing employees and employing only a few designers. Now, they can produce a lot of value with only a few cost inputs, hence productivity just went up. Of course, the designers may not be more productive in terms of creating more designs or better designs, but as Apple is able to capture a large share of the total value, then it will be attributed to the super productivity of the designers, which may have more to do with politics in East Asia choosing an export led growth model than anything to do with the iPhone.

And it gets more complicated because the NIPA system doesn't take into account the possibility of economic rents. E.g. going back to the Apple example, if Apple can charge $1000 for the phone, with only $300 of inputs, then this $700 worth of "value added" must be attributable to some portion of labor and capital. If the designers are paid $200 then capital is paid $500. In NIPA parlance, that means $500 worth of capital investment (and or depreciation) must have occurred somewhere in the economy, which again to the layman would require unrealistic measurements of investment. There is no NIPA notion of someone overpaying. And then don't get me started on how corporate shell companies headquartered in tax shelters like the Bahamas or Ireland screw with GDP numbers, especially of the smaller nations that serve as tax shelters, but also with US productivity and investment data.

Especially in the context of Intel using overseas fabs, the dire bit of that article is

> a lot of the manufacturing output growth since the mid-1980s, when the BEA started factoring quality improvements in computers into its measures of real output, has been driven by the rising quality of computers and electronic products, not by increases in actual, you know, output.

i.e. one Intel chip produced today is considered to be thousands of times more output than one Intel chip produced in the 80s, because it's thousands of times faster.

> i.e. one Intel chip produced today is considered to be thousands of times more output than one Intel chip produced in the 80s, because it's thousands of times faster.

This point is a bit of a rorschach... is that an over-estimation or an under-estimation? Sounds like a massive under-estimation to me, in the same sort of qualitative way that it's silly to try to quantify the economic effect of the internal combustion engine in terms of horses displaced... it's measuring something kinda relevant but totally missing a qualitative shift that unlocked impossible-to-quantify economic activity (for better or worse is up to us).

Sure, but remember where we started - an argument about whether there is more or less domestic manufacturing than in the past or not. Increasing the performance of the product does not mean more manufacturing is going on - that mistakes value for output.

> that mistakes value for output.

Yes, exactly. My point is that I wouldn't call measuring actual value instead of a pretty meaningless "output" metric a "mistake".

The original point of the pre-existing "output" metric was to approximate actual value creation, after all...

Perhaps another way to look at it would be that, as the quality of computers has increased, the total number of computers being made worldwide has also massively increased.

If we treat each computer made in the US today as equivalent to a thousand 1980s computers, then I'd hope to see the weighted data show that US manufacturing has massively increased as a result of making an exponentially larger number of exponentially faster computers. Rather than seeing it stagnate despite weighting each computer as worth so many old computers.

> I'd hope to see the weighted data show that US manufacturing has massively increased as a result of making an exponentially larger number of exponentially faster computers

Why would you hope for that?

That's what I'd hope to see as evidence to support a thesis that American manufacturing is not declining.

Thanks! Bloomberg has become one of my favorite sources for real news.

Like spy chips on supermicro motherboards?

After that story, they’ve lost a good deal of credibility.

I find it hard to believe that such a well informed crowd with health dose of skepticism as HN can be outright dismissal of that article just because no evidence has been revealed, and because experts have dismissed the story so far.

A favorite story on HN, this listening device was not detected for seven years, despite experts extensively looking for it [0]

[0] - https://en.wikipedia.org/wiki/The_Thing_(listening_device)

As a side note, we tend to judge an article if it goes against our belief, but are willing to give it a pass otherwise. Case in point, plenty of Bloomberg's articles about Boeing or climate change being discussed but don't think I've seen anyone questioning their premise or evidence. I'm guilty of this as well.

If you’re going to make a claim like that, you need to make it in a way that experts can verify: absent follow-up or reporting from other, independent, outlets, there’s no good reason for me to take it seriously. And, given Supermicro’s connections with Taiwan, there’s plenty of reason for me to suspect something’s up.

I wonder if someone planted that story just to provide something for people to complain about.

Bloomberg could choose to start earning back some trust at their leisure by retracting the article.

They haven't; they still stand by it. Since they consider the rest of their output just as trustworthy as that article, I'll take them at their word.

The FT was under intense pressure for years to retract their stories about Wirecard. Journalists even got sued by the German regulator. And most disagreed with FT enough that stock prices remained high for years. They never retracted a single article and trusted their sources and journalists more than the public.

In the end they turned out to be right but it took more than 5 years since the first article. I believe we should consider that Bloomberg could fall into a similar category.

Tell that to the industrial midwest. If you look at statistics of # factories in the US over time, in every size category, the number of factories is going down.

Chip manufacturing is a special case as well. China is using massive state spending to build a competitive infrastructure in its country. Intel is the only competetive industrial chip fabrication presence in the US and should receive a similar focus from our leadership if we don't want to fully hollow-out the US' production capacity.

This is too simplistic.

Samsung and Globalfoundires have fabs in the US that are close to leading edge. There are other fabs that are leading edge for other types of chips e.g. Texas Instruments

The fact is that the capital expenditure costs to create a leading edge Fab are insane, and economies of scale form a big moat. The economics do not really support more than a few leading edge fabs, and now that Intel has had internal issues they are behind. Maybe they will catch up- however throwing money at the problem does not seem likely to me to fix the problem. Moore's law is ending and the difference between die shrinks isn't as dramatic as it used to be. It remains to be seen if electron tunnelling can be easily be dealt with... However fabs are sexy and people will chase the money until the process breaks down.

So we should just give up then and hand the IP to China? I do not understand this way of thinking Intel became Intel because of innovations they did.

Technically we’re selling it to China, not giving it. Notice:


Intel are assholes enough, commercially, now. How do you think they would act with the US government's support?

That would actually be an anti consumer move.

Do you mean throwing public money at a private company? How is that even doable in a capitalist country and why are Intel shareholders the ones blessed to receive free capital influx?

Like I said, the US is unique in its mass deindustrialization. Somehow Germany, Japan, and South Korea have avoided it. Last I checked, they are capitalist countries as well.

Directly throwing money at a company isn't the only way for the state to support an industry. I'm not a policy wonk, but there are other ways to do it, and I think the US would be making a mistake if they let this chip fabrication capacity disappear.

Not really unique, Great Britain is another example, although the underlying reasons might be somewhat different.

Intel's problem isn't with money. Intel is a hugely profitable company last time I checked. And It definitely has a lot more money than its challengers, receiving government fund or not. It's quite funny we are so used to this kind of capitalist view that thinks money solves all our issue.

That's the very problem. It is not lack of money that is making Intel to lose ground, it is simply too easy for them to send checks to their investors and give millionaire bonuses to their executives. They're following the footsteps of IBM, GE, and Oracle.

> It's quite funny we are so used to this kind of capitalist view that thinks money solves all our issue.

This is silly. No one thinks that. You're just not telling the difference between "necessary" and "sufficient".

> Somehow Germany, Japan, and South Korea have avoided it. Last I checked, they are capitalist countries as well.

Japan and South Korea are not really; their economies are largely organised by conglomerates with strong political ties (and often significant nationalistic tendencies). There is capitalism in Germany but it's kept on a much tighter leash than the US in terms of labour laws.

This is accurate. SK has Chaebol and Japan has Keiretsu which has revolving doors to technocrat bureaucracies and governments where jobs, power and money are free flowing between these entities.

This is historical fact, not a theory.

South Korea is very reliant on China for manufacturing, and I wouldn't be surprised if it's even more than the US considering the close proximity of China and the lack of natural resources in Korea. It's generally acknowledged by Koreans that the most abundant resource in Korea is human brain power, especially with the high college graduation rate (50%+, versus 30%+ in the US). The stigma of manual labor jobs such as manufacturing and agriculture are very strong, leading to a rapidly aging workforce in those sectors, and this trend, like in many other post-industrial nations, continues to increase.

The US does plenty of throwing public money at private companies. Certain industries (banking, airlines, energy, agriculture) seem to show up at the trough on a fairly regular basis for their turn to be bailed out.

National security. See Raytheon, General Dynamics, Northrupp Grumman, Lockheed Martin, Boeing, etc...

To the detriment of humans elsewhere suffering the consequences of those weapons

Like the burgers you flip that feed soldiers that shoot? The capitalist chain reaches every one. You are equally guilty then? Nonsense.

I like this. We are all culpable.

Every single human that spends a single dollar is directly and demonstrably connected to systems that are rapidly (in historical proportions) destroying almost everything that humans of past eras considered wealth and value.

Mother earth.

Downvote this, but let it sink in.

There are humans on this planet that are not fueling these consumptive mechanisms. I hope.

How can we detach ourselves from these systems as we build out alternative systems?

If we reframe the problem as: How do we run an economy not based around GDP growth, ie. Not based around birthing more people to increase labour? Japan is ahead of the curve here in that it has an ageing population, they appear to have placed their bet on robotics.

Fundamentally this is an energy issue, with enough free energy everyone can have a high standard of living ie. BI becomes possible, we can automate away most mundane jobs with cheap robots (BOM goes down with cheap energy).

With a good standard of living, births drop- viz population charts globally except Africa, thus we get into a virtuous cycle.

The only issue is the environmental impact of lots of energy in the ecosystem, I'm not sure how that would pan out.

I don't buy this.

We are wasting massive amounts of energy.

20 lanes of stop-and-go SUV's all single driver traveling over an hour to jobs that they hate. Terrajules of energy to create a single piece of electronics designed to break in less than 2 years. Massive manufacturing of throw away disposable items. The list goes on.

This is Hacker news. Que's Computer Users Dictionary 5th edition defines hack:

"An inordinately clever rearrangement of existing system resources that results, as if by magic, in a stunning improvement in system performance - or an equally stunning prank."

We need to wisely use what we already have.

Build another one, then the story goes that you have to build weapons to defend it against "the old systems", as now the cake is all divided and such, and there is no new world to discover on earth but in space.

Who make's their chips?

That is basically what happened in the 1980s in the US with SEMATECH. I remember a computer tv show in the 80s talking about government participation. I didn’t hear qualms about it then.

I think it was fear of lost of the memory business to Japanese companies. It worked out for the US in that Intel switch to and dominated the CPU business and you had the rise of fabless. .

See https://issues.org/van_atta/

The article is very interesting in describing the origin of EUV and the company making it and the source of the research.

> Do you mean throwing public money at a private company? How is that even doable in a capitalist country

Lockheed Martin. Boeing. Raytheon. General Dynamics. Northrop Grumman. Huntington Ingalls. United Technologies.


Or assess tariffs on foreign imports from nations that don't compete under the same labor and environmental rules as the USA.

That would likely make matters worse, not better. Placing tariffs on goods from SE Asian countries (including, but not limited to, China) would increase the cost of manufacturing many products in the US which rely on materials from those countries, making those products less competitive in foreign markets.

Would you open a manufacturing plant in a location where people don't agree to anything, not even wearing a mask that is in their own best interest? I would not.

"Manufacturing" as a business classification sometimes seems to include companies that are in the business of developing and selling manufactured products but don't own or operate any factories themselves and outsource the actual, well, manufacturing. In the semiconductor industry people talk about "fabless" companies as a category but there's not really an analogous shorthand for, say, a factory-less soap manufacturer.

With that said, I wish stats like that were a little more clear about whether they mean that broader definition of the manufacturing industry, or if it is strictly the economic output of actual factories.

People love quoting this, but you’re just cherry picking.

This has quite thoroughly been criticised in the past, repeatedly, for the way the data is adjusted.

this image contradicts everything I've maybe ever heard about "American manufacturing"'s decline. Not saying the image is wrong, it's just impressive how much one segment of industry has been politicized.

It's important to note that the number of manufacturing jobs has fallen significantly. Manufacturing has gotten way more efficient since the 40s, so a lot fewer people are necessary.

That's why it's so politicized. Manufacturing is about 12% of GDP. That 12% used to go to a lot of people. Now it goes to the few people who own all the factories and the robots inside them.

> Manufacturing has gotten way more efficient since the 40s,

It's both yes or no for USA:

1. Fever businesses at all in the sector

2. USA has an extremely strong slant towards expensive capital goods, in very entrenchable industries: aerospace, defence, weapons, medical, industrial vehicles, and machinery

3. All of the above is increasingly made with imported, off the shelf parts

4. USA has "industrial" companies who manage to report giant manufacturing outputs in accounting statements, but known to not have a single running manufacturing plant.

Do you have an example of 4?

I don't think its quite that extreme, but Boeing, car companies etc all extremely large amounts of subcontracting.

thanks, that's important to note and definitely explains why it is politicized.

The US decided to get rid of all manufacturing that employed lots of people, so politically they would destroy the power of unions. Unlike other countries, American politicians have no shame about it.

Also US companies were just cutting costs- de-powering unions was a happy bonus. Technology allowed it (transportation, communications and cheap energy). But also the government allowed it, because it's all controlled by corporations.

> it's just impressive how much one segment of industry has been politicized

Job losses in manufacturing have sent millions of people from middle class to poverty in the span of a single generation.

Manufacturing is also directly tied to national security, regardless of how many choose to ignore this.

imo it hasn't been politicized enough.

I imagine a lot of it is due to automation - there are in fact less assembly line jobs than there were in the 40s but economically the sector has continued to grow.

Automation in the industry is a big lie, the degree of it since post-WW2 days (when most of modern mass production methods were perfected) is greatly overstated.

It is mostly due to outsourcing the labor-intensive parts to cheaper labor markets.

The hypothetical robots are people working elsewhere.

> it seems the US is almost unique in our mass deindustrialization. Japan, Germany and South Korea have avoided this.

This is very common misconception. The US has not deindustrialized other sectors have grown faster. US is the second largest country by industrial output. Third largest if you count EU as comparable unit.

Industrial era is over in all advanced economies. We live in post-industrial societies.

It's not in China, Germany, Japan, or South Korea. They still have a huge manufcaturing presence. Samsung builds its products in South Korea. Everything has to get manufactured somewhere and, in my view, the US has made a mistake by outsourcing this capacity more than other countries.

As I said elsewhere in the thread, chip manufacturing is a special category as well, and Intel has the only US-based fabrication that is competitive with the state of the art.

It should be as important to protect that capacity as other industrial sectors, like Boeing.

Yeah, China is a manufacturing powerhouse. But the world system has moved beyond that. The US can't compete on price.

Germany and Japan have stagnant economies. South Korea and the US grow at similar rates but the US has a massive base to grow on.

America's economy grows at 2.9% every year. The largest in the world and it's still growing at this rate.

I take this trade every single time. It's why Americans are so wealthy in general.

Are you talking about the same americans that required huge bailouts from the government the second the economy stopped because 4 out of 5 live paycheck to paycheck?

That's right. The same Americans who are nonetheless more wealthy in the median than most people in Europe or anywhere else outside North America.

Outsourcing your safety net to the government is the classic move of a well-functioning society.

If you are living on credit in what sense are you wealthier?

> Outsourcing your safety net to the government is the classic move of a well-functioning society.

Are you watching the same news as the rest of us? where is this well-functioning society you speak of?

>Are you talking about the same americans that required huge bailouts from the government the second the economy stopped because 4 out of 5 live paycheck to paycheck?

Yeah, it's their fault for not expecting and being prepared for the government banning the majority of commerce.

If 4/5 Americans cannot stay afloat 1 month on their own then yes, it is a big problem. If someone gets into an accident, or has a medical issue, or suddenly loses their job, it puts them into the same position.

I think we should arrive for a society where it's citizens can earn enough on their own to at least have a small safety net.

It is a problem but it raises a question: Are 4 / 5 Americans unable to stay afloat because they are underpaid or because they expand their expenses to fit their income?

We talk about a race to the bottom irt. companies but it might be wise to apply the same thinking to society at large as well.

Companies have been criticized(rightly imho) over their lack of reserves to handle the corona virus and shutdown. The defense of the companies is that by keeping cash on hand instead of spending it, they are at a competitive disadvantage.

This holds true for individuals as well, abet in a more abstract manner. If you are young and trying to attract a mate, wearing old/unfashionable clothes is detrimental. In some circles, what car you drive and it's age is important.

This thinking also extends to non-tangible things as well. What TV shows do you watch? Subscription to cable or even streaming is expensive but it is something we often feel we have to have, not only for our entertainment but also to be connected culturally with our peer group.

These pressures and factors are not new, the phrase "Keeping up with the Joneses" came from a comic in 1913. It has been argued that social media is making these pressures even greater, as people share(and reward the sharing of) the positive aspects of lives but leave out the negative aspects. This creates a sort of perverse view of the world where you see people taking vacations, going out to eat and buying new gadgets where it all runs together and does not show the effort, hardships and sacrifices made to obtain those things.

If we as a society do not value having savings then why should we expect anyone to save? Especially is the reward for doing so is not visible?

> Samsung builds its products in South Korea

Actually no, they build phones in Vietnam and India. They just avoid China.

Samsung's bread and butter, components (eg, dram and displays) business are still made in South Korea. The company also recently announced that they are investing $100B over next 10 years in their foundry business, which is likely to be all in South Korea.

My Samsung phone says Made in Korea on the back. I assume that means it was assembled there and mostly uses Chinese components on the inside like everyone else. It is almost impossible to avoid China if you are in the manufacturing business.

I highly doubt your Samsung phone contains many Chinese parts. The camera, screen, chip come from Japan, SK, and Taiwan. Maybe the speakers were made in China.

I recently read this review https://phenomenalworld.org/reviews/trade-wars book sounds quite good as a sober explanation without the blinders of neoliberalism or industrial romanticism.

And it would be even more massive if we actually made anything. In fact if it weren't shrinking by volume it might be growing faster than those other sectors considering how much of the economy is "consumer" driven.

Go to the mall/shopping center, and try to find something made in the US that isn't a bulk chemical product, or something so low value in relation to its weight that its still made in the US (paper). If you go back ~30 years it was the complete opposite. If you found something manufactured overseas it was unusual. So in just a couple decades we went from pretty much 100% of consumer electronics, clothing, furniture, general home appliances, cookware, toys, etc made in the US to a tiny fraction. Even in areas where we still manufacture things, its rare that 100% of the products in a particular segment are produced here. Sure, you can buy American made tires, AC units, etc, but most of the volume isn't made here. But even in the case of AC units its frequently assembled of foreign made compressors/etc where the 4 bolts, two pipes and 4 wires that need to be connected (by $15/hour labor in basically warehouses) get more value add than the actual hi tech parts made in high capex factories. All for the purposes of computing "made in America".

> Go to the mall/shopping center, and try to find something made in the US that isn't a bulk chemical product

Please don't use anecdotes and personal experiences when discussing economics.

You lament about US not manufacturing low value added products. You can't have personal experience that would give you a good view.

You must use statistics and data to argue.

Fine, lets play that game.

Show me the statistic that say how many shoes/sandals/boots were sold in the US last year. Now show me what percentage of those shoes were made in the US of foreign components and what percentage of them were made beginning to end of US sourced components.

Many of these companies can't even keep their internal supply chain logistics in order enough to report what country something was actually manufactured in, much less reporting hard volume/etc numbers sufficiently to do anything but obfuscate.

Put another way, if the "economic" numbers weren't so cooked they might be a reasonable basis for argument, but time and time again when you dig into something you quickly hit a dead end because no one actually collects such detailed statistics across the entire economy. Sure you can tell me how many planes Boeing made in each of the past 30 years, but good luck as you move down the supply chain, or for less valueable goods like refrigerators.

If you look at the computer industry you frequently can't even tell how many units a company sold in the US because they want to obfuscate their declining volume. To be more specific, can you tell me how many zseries mainframes IBM sold each year for the last 10 years? The last couple times I looked at this, the closest you can get is "installed capacity" which is IBM multiplying the machines by their theoretical capacity which is measured using some proprietary IBM metric and doesn't even represent actual sales because they have "capacity on demand" which allows them to sell/install a machine with capacity X but only charge and provide the customer with a fraction of X. So, each generation of machines is some larger multiple of X, which allows them to sell fewer machines (and capacity), and claim more installed capacity.

You will get a more realistic view going to your local target, and standing in the sock isle counting what people are buying than you will get from the target sales numbers.

(socks are another one of those areas where its actually possible to buy US manufactured textiles if you look, because they are the kind of garment that is easily automated. Its likely even US cotton, but whether it was shipped overseas for processing first, good luck..)

China is not post-industrial and without their factories US businesses would have no products to sell. China can change its mind about cooperating or drastically raise prices any time they want. The US has put itself in a very precarious position by dumping all manufacturing. Manufacturing is what adds real value to an economy. Without something to sell it's turtles all the way down. OTOH US labor is not competitive and that is a very real problem. Americans have sort of become too rich for their own good. At some point every single person has to start their own business because being an employee will not pay enough to keep up with the cost of living.

> Manufacturing is what adds real value to an economy.

This is yet another misconception that only manufacturing is real.

Share of services as the value of manufactured products is also increasing. In developed countries it's usually 40-50%.

China produces low value added products. Producing what China does in the US would reduce value added in the US economy. The US would become poorer.

Manufacturing is the foundation upon which all tertiary and quaternary industries are ultimately built.

It is pretty clear that you are not using the same definition of value as the person you are responding to.

It's not great to be #2 when #1 is run on a commitment to oppose human rights.

It's "deindustrialization" when US has slid down from "#1 by far" to "#3".

Yes because the we don’t do things like blow up other countries to “bring democracy”. We also don’t have anywhere near the highest rate of incarceration in the industrial world. We aren’t the only democratic nation that still has capital punishment and we believe that health care should be a universal right.

Not to mention we would never have the federal government sending a police force into cities that don’t want them.

The US is not exactly a “light on a shining hill”.

Fascinating way to look for deindustrialization. In this method, if China had some kind of super-SARS that killed everyone, then the US would have reindustrialized!

> Interestingly, it seems the US is almost unique in our mass deindustrialization

The UK has vast swathes of poverty in the deindustrialised north of England and other areas. Former mining and heavy industry towns that even today almost 40 years on from deindustrialisation are dominated by insecure, unskilled work and unemployment

We really, really, really NEED US based fabs.

>US is almost unique in our mass deindustrialization. Japan, Germany and South Korea have avoided this.

yup, I wouldn't necessarily include SK in that list but I'm very displeased at our deindustrialization process. Most/all of it does not make sense outside a tenuous abstract global finance system.

We are losing knowledge and control every time and every day factories are allowed to leave the country.

But running a fab is a huge cost and liability while all the competition is outsourcing to TSMC. Because the process is so difficult, I think it makes sense to only have fabs that work for multiple companies. I do think US chip companies (Intel, AMD, Nvidia) and government should get together and start a fab by themselves.

The reason market challengers outsource to TSMC is because they don't have enough revenue to support their own fab line, with Intel taking majority of the revenue. The precise reason Intel kept their own fab is because it's the makret leader that wants to keep their competitive edge, that is under the assumption their own fab is more advance, and for their use only. However, now that TSMC as a fab somehow got more advanced. It meant Intel's in-house fab has failed in that mission. That doesn't mean Intel's challenger's gonna hand their production to Intel's fab. Intel obviously won't just turn over their technology for their competitors to use either. It's quite impossible and unconventional to do.

> it seems the US is almost unique in our mass deindustrialization. Japan, Germany and South Korea have avoided this.

Entire Central Europe (Poland, Slovakia, Hungary etc.) is filled with German manufacturing plants. The Germans have done plenty of deindustrialization themselves.

47% of German GDP is manufacturing. Hardly deindustrialisation.


I wonder how is that counted - i.e. can for example Mercedes contribute a lot towards the part of GDP counted as production even if most of its cars are produced outside Germany. Or, is that counted as engineering and branding.

>Interestingly, it seems the US is almost unique in our mass deindustrialization. Japan, Germany and South Korea have avoided this.

Related to the dollar being global reserve currency, which encourages a trade deficit. The US has a deficit equal to the next 15 biggest deficits combined. On the other hand, Germany, Japan and South Korea are top 5 trade surpluses.

The US hasn't deindustrialized at all. Industrial production has increased pretty consistently as far back as I can find data.

What has increased in pace, though, is capacity per person, so manufacturing _jobs_ have decreased. It's less of a geopolitical issue and more of a domestic employment issue.



Ironically, those numbers were cooked up by counting Intel as manufacturing and over-weighting its success with a hedonic multiplier:


Idk what a hedonic multiplier is, but my understanding of that article is that production numbers measure the real $ value of output, not the actual quantity, making the production index misleading (e.g., if the US manufactured totally stopped except to make 1 computer chip costing $100 trillion, then manufacturing would seem to shoot up, even as it actually collapsed.)

If that's correct, then it's interesting: we're making few, valuable items now, rather than many, cheap items. It follows that automation isn't a drive towards cheaper production, but more a drive towards higher quality goods that simply require machines to make.

But it all takes me back to my point: from a geopolitical perspective, being good at highly-differentiated, tech-intensive, difficult-to-execute manufacturing is a great position. It's just a disaster from an employment perspective, because to get there, we traded away all our labor-intensive manufacturing.

Hedonic multiplier is the idea that a CPU that's twice as good should count for twice the GDP even if it doesn't actually sell for that. It's a strategy for overweighting our strengths in those manufacturing indices by confusing the relationship between money and value.

Intel was the lynchpin of this lie. We are in a thread about Intel becoming non-competitive at highly-differentiated, tech-intensive, difficult-to-execute manufacturing.

No, that is not a great geopolitical position to be in.

Yes. Intel's failure demonstrates that the US no longer knows how to make chips as good as those of our geopolitical rivals. We're kept safe in large part by our technological edge. If we start falling behind in technology, it's not just a bad decade for Intel, but a hole in national security.

Taiwan and South Korea are hardly our geopolitical rivals.

Oh, come on. We both know that in a war, our main geopolitical rival would conquer both on turn 1.

We really don't. Naval invasions are hard, particularly when the US is the world's dominant sea power. Theoretically China could invade South Korea through North Korea but North Korea wouldn't like that and they have nukes. But conquest of them is what happens when WWIII cooks off - we were talking about geopolitical rivalry.

Even if China can't invade Taiwan right away, China can set up a naval and air blockade. And if they can't do that, they can destroy the very delicate fabs using short-range conventional missiles. China is a high-tech peer adversary, not a bunch of desert guys in pickup trucks. As a matter of national security, we need a secure (and E2E-trusted!) supply chain for advanced technology.

Another article I just read says Intel is talking to TSMC. I think they are trying to get out of manufacturing.

While I'm not a fan of Intel's monopolistic position on x86, but if Intel really gets rid of manufacturing, it's going to be another accelerator of consolidation in the semiconductor industry. I see it as a threat and I hope it will not happen. The timeline which Intel's in-house fabs survive, is a better timeline.

Samsung has pledged to heavily invest in fabs and try to beat TSMC while SMIC is gaining historic momentum too. Looks like good competition in the upcoming decade to me.

A monopoly is not defined as you’re the only one that can make your own products.

Intel doesn’t have a “monopoly” in the processor space.

It is going to be a little bit geopolitically terrifying if Taiwan/TSMC monopolizes cutting edge chip fab.

TSMC is building a $12B factory in Arizona.

It will never be the latest chips. TSMC will never let their cutting edge fab machines out of Taiwan. If they're on 3nm then Arizona will make 7nm or something.

Except those machines are largely built by other companies, like ASML...

Built by, but the manufacturing processes and know how to design them (EUV) are TSMC. Good point, it's more complex than just TSMC :)

We'll see. Bargains come in many forms, and Taiwan certainly has things they want from the US.

Like what ? F16s ?

Aircraft carriers and nuclear ICBMs. F-16s and missiles are merely to buy enough time for the U.S. to respond, which in a surprise attack with a minimal initial invasion force probably doesn't require as much sophisticated equipment as Taiwan has to preserve a viable defensive line for if and when the U.S. intervenes.

If China (PRC) was smart, it'd have invaded Taiwan two years ago as Trump's resolve to backup commitments to Taiwan, or pretty much any military commitments to any ally, with force is basically nil. Nearing the election it's less clear as retaliation and a standoff crisis might better suit his reelection bid under present circumstances (e.g. COVID-19 sentiment).

China missed its opportunity, but if Trump were to win re-election China would have much less to lose given their rejection of one country, two systems, the deterioration of trade ties, etc. In a lame duck Trump presidency invasion would have an even better cost+benefit profile for China.

China doesn't need to invade Taiwan. It just need to create this pretext that they will to prevent Taiwan from becoming a US military base.

Geopolitically I agree that there's little justification for China to invade Taiwan. But AFAIU there's a very strong cultural mentality, especially among senior military leadership, that is absolutely obsessed with Taiwan. The fact that the annexation of Taiwan provides almost no substantive benefits whatsoever is presumably all that holds them back, given the slight risk of a U.S. response. Growing Taiwanese industrial prowess and the benefits that inure to the U.S., such as with TSMC, upsets the precarious balance by making Chinese control of Taiwan strategically more important.

It's the same thing with South Korea. The U.S. would likely agree to leave the Korean peninsula in a heartbeat if China agreed to regime change in North Korea, even if it was kept as a buffer state. Geopolitically it would be better for everybody involved. But senior Chinese military leadership feels honor bound as the protector of the North Korean regime, which also stands as living evidence of Chinese military superiority over the U.S. And so China defends the most egregious behaviors of North Korea, even when they openly defy China on the international stage, and even when North Korean policies cause domestic trouble in China.

A comparable mentality in the U.S. might be that regarding Cuba--completely and utterly irrational, driven by an antiquated historical narrative and unjustifiable sentiments. In some respects an even better example might be Israel, but it's exactly because it's an apt comparison that I can't actually make it--there's little room for nuance and too little good faith allowance in the domestic discourse.

Why? The only part of this that worries me is how China might react to it.


Yes, I myself heard that Intel CEO been frequenting visits to Hsinchu last year.

There is only one company out there worth visiting in person for a man of his calibre.

Can you elaborate?

I just learned about https://en.m.wikipedia.org/wiki/Hsinchu_Science_Park

Would this be to acquire, poach, invest, or relocate?

The biggest company there is TSMC.

The murmur been that Intel gave up on 10nm, and wanted to do a tapeout at TSMC's fab as a stopgap measure until they can make a new process from scratch.

Maybe for Altera FPGAs? I don't really understand the scale Intel needs and whether TSMC offers enough for their micros.

> Interestingly, it seems the US is almost unique in our mass deindustrialization. Japan, Germany and South Korea have avoided this.

Those countries also have much higher trade barriers. The US is almost unique in how few trade trade restrictions they (used to) have.

>Those countries also have much higher trade barriers.

This isn't true. Here's a comparison of the EU, several individual European countries, Japan and the US[1], and there isn't much difference tariffs or trade barriers.

Global trade-barriers have been reduced to virtually nothing among developed economies, the US isn't unique at all. This is a stale old meme that seems to be popular in the US because it frames the US as a sort of victim of global trade which is a popular political narrative but not evidence backed.

The real difference is in the overfinancialised, laissez-faire capitalism of the anglosphere (the UK suffers from the same issue of deindustrialisation, Australia as well), comopared to the industrial policy and state/big business dominated capitalism of Japan, Germany, France and so on.


There is of little significance advocating a company to do that. Unless it's regulated and legislated. Any company who willingly gave up such advantange in cost would be a fool and lost its market due to such policies unless that company has other unique advantange to keep their market position. People like to talk about how they love "locally-sourced products" until they realized they had to pay double or triple.

Not that unique.

UK did it as well, we let Arm be sold to foreign investors - I wonder if the French or German government would have allowed that sale.

Arm is hardly a company that manufactures anything. An impressive tech company for sure, but I don't understand how the foreign takeover of a company that licenses processor designs is an example of loss of British manufacturing.

Just spent 3 months in Hillsboro, Oregon, and Intel construction work there is full speed ahead it seems. Of course that could change pretty quickly, but right now it sure looks like they intend to keep fabs going in the US.

> Interestingly, it seems the US is almost unique in our mass deindustrialization.

The UK, Belgium did the same. The whole Communist Bloc also did it (for different reasons).

France too.

> make a lot of short term profit (cut expenses)

> invest into long term stability

Well, you have to choose one.

half of chip making is in manufacturing it. If they out source it, not much would be left.

This is clearly false. None of Nvidia, AMD, or Qualcomm manufacture their own chips. These companies are valuable despite not having in-house manufactures, perhaps because of it.

> half of chip making is in manufacturing it.

Correctness of this statement also depends on the type of chips.

Not everything needs the latest and greatest process node (which is predominately a fabless industry) except for advanced microprocessors, SoCs, or storage. On the other hand, 65 nm is perfectly fine for 90% of chips in existence, even 0.18 microns is still often used. You don't need EUV for a good LED driver, a serial port, or a battery charger controller. If we talk in this context, indeed, companies like Texas Instruments, NXP, ONSemi, Analog Devices, Microchip, etc, usually make chips in-house.

They dont have to outsource their Core Asset. If Intel ships 250M CPU annually there are also 250M Chipset going alongside with it. Those can go to TSMC or Samsung. ( Or in a turn of an event, Global Foundry )

> I hope Intel keeps its US-based fabs and continues to upgrade them. It would be a mistake to cede this capacity to outsourcing like so much of our other manufacturing capacity.

Intel is a publicly traded internationally-owned corporation, with no obligations to the long-term welfare of the United States.

If you'd like a chip manufacturer to make decisions based on what is good for the US, I would recommend bribing it, nationalizing it, or starting a government-backed competitor.

Any of those would be more effective at accomplishing this than hope and prayers. The free market itself is not necessarilly going to deliver the results you wish for, just because you wish for them.

Gee save us the free market lecture friend. What he said doesn’t exclude any of the options you’ve brought up.

Also I note you’ve raised carrot strategies when stick strategies exist also.

You are correct about having no obligations, but as a company that is based in US and employs a large US-based workforce it benefits a lot from US itself being successful.

In Today's World, Countries should be grateful to a company when they employ people. There is no lack of highly skilled workforce in entire world. US with it's abysmal education system should be grateful to intel that it attracts semi conductor industry talents from all over the world which maintains US's hegemony in the world. If intel doesn't employ them, they would start companies like TSMC in their countries, hell if US didn't discriminate against Asians in 1980s maybe TSMC wouldn't even exist in the first place. Benefits that US provides to companies is way less than benefits companies provide to US.

Applications are open for YC Winter 2022

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