Sharpie Found a Way to Make Pens More Cheaply by Manufacturing Them in the U.s.
Key topics
Sharpie has reduced production costs by manufacturing pens in the US, but commenters question whether this is a genuine success story or just a result of automation and tariffs, highlighting the complexities of US manufacturing.
Snapshot generated from the HN discussion
Discussion Activity
Very active discussionFirst comment
52s
Peak period
94
36-48h
Avg / period
14.5
Based on 160 loaded comments
Key moments
- 01Story posted
Oct 5, 2025 at 6:01 AM EDT
3 months ago
Step 01 - 02First comment
Oct 5, 2025 at 6:01 AM EDT
52s after posting
Step 02 - 03Peak activity
94 comments in 36-48h
Hottest window of the conversation
Step 03 - 04Latest activity
Oct 13, 2025 at 2:52 PM EDT
3 months ago
Step 04
Generating AI Summary...
Analyzing up to 500 comments to identify key contributors and discussion patterns
Want the full context?
Jump to the original sources
Read the primary article or dive into the live Hacker News thread when you're ready.
https://www.cnn.com/2025/08/19/economy/us-tariff-rebate-chec...
Seems closer to plugging half the gap than "cancelled"
And oh, is this also with that made up math they used to say that continuing the existing tax cuts past their expiration date didn’t count as increasing the deficit?
Wealth is the stuff - in this case pens. Money is what you get in exchange.
Society gets more pens with less human effort required - that's a win, and has nothing to do with money.
Money (increased profits, share price, etc) is what happens when you create the more stuff with less effort that society gets.
Money is how we keep track of how much net wealth (stuff) someone or some company has created for society.
Marketing this as a success story of U.S. manufacturing is insane. If the WSJ honestly thinks the outlook is better now, it should at least provide the history and say why (and who provided the investments for automation given that Trump is a sharpie user).
There’s a list of nice business steps the company took (and I can’t imagine starting work on the problem in 2018 hurt either), but I don’t expect they were the only ones to take any specific one, so why did Sharpie in particular succeed? What’s the recipe? Automate the crap out of your assembly line and promote (a lucky few among) your former assembly-line workers (who you definitely did not fire when you got high on automation) to technicians? I can’t imagine that’s a rare thought; yet this seems to be a rare success story.
The most newsworthy detail here is probably the WSJ publishing an article that could be construed as somewhat pro-labor.
> Peterson […] found that the factory could use robots to do an increasing share of the packing. But he decided to keep the employees who knew the company and convert their jobs to roles such as automation engineering. In that case, an employee would fix a robot instead of packing a box. Peterson estimates the average wage at its Maryville facility, which employs 550 staff, has gone up some 50% over the past five years—without a change in head count.
Like a coder writing code (by hand!) to make things more efficient somewhere, also seems noteworthy now.
https://en.wikipedia.org/wiki/Manufacturing
17%+ with 4% of the world’s population means there must be some kind of automation expertise.
Which has more industrial might? You need to compare what's actually being made and how much to draw conclusions.
Skills like manual machining, and manual welding have gotten much harder to find and a smaller portion of the economy.
As to if that matters?
What Sharpie did in the US is super simple table stakes for manufacturing anywhere, it’s only notable because it can be used in a larger political story.
Before Boeing had its epic fails, usually they were talked about as the prototypical “we make airplanes instead of rubber dog shit out of Hong Kong” example.
But it neglected to show that Boeing outsourced a huge portion of the actual skillsets needed to build each individual component.
There is utterly no comparison to $100B of airplanes being sold a year vs. $100B of various goods of diverse complexity and quality in terms of impact to the economy, workers, manufacturing knowledge, and even national security.
If we had simply gone up market and retained useful engineering and manufacturing skills/talent pipeline/capacity I’d totally agree. But we have not. Not to any appreciable degree by any metric other than numbers on a spreadsheet.
We are talking about this in a thread discussing manufacturing markers in the US as some sort of large win. If this article had been discussed in these contexts 60 years ago it’d have been seen as utterly pathetic.
> It is not from the benevolence of the butcher, the brewer, or the baker, that we expect our dinner, but from their regard to their own interest.
The Model S was an expensive car with a bad reputation for quality. Then in 2017 they introduced the Model 3 and their "Alien Dreadnought" automation. That significantly delayed mass production of the Model 3 and almost bankrupted the company.
Tesla turned things around by building a factory in Shanghai, and learning how to build a car from the Chinese. They then basically copied the Shanghai factory to Germany.
Fremont factory has produced 3 million cars, in a manufacturing hostile/barren California (and thus Nevada gigafactory), covering ~50% cars produced yearly. [1]
> Tesla turned things around by building a factory in Shanghai
~50% are made in China, with ~50% of those being sold in China (well until this year), avoiding significant Chinese tariffs. [2]
[1] https://www.teslarati.com/tesla-fremont-nevada-3-million/
[2] Nice historic graph, including poor sales this year. https://carnewschina.com/2025/03/10/tesla-exports-3911-cars-...
Some of their original manufacturing ideas didn't work out, and they iterated to better processes. That's not failure. That's plain old boring fucking engineering.
Do you think China is bad or something? Everyone, especially in the industry, knows that there's more manufacturing expertise in China, because that's where the vast majority of the world has outsourced their manufacturing (and pollution), for decades, and the forseable future. Just ask the politicians, who have been singing "manufacturing is never coming back to the US" for decades, to justify their policies that prevent people from manufacturing in their backyards.
Who learned Kaizen from an American.
Good ideas are useful to successful people.
It is not, why would you think such a thing?
https://nam.org/mfgdata/
I'm somewhat surprised that the other response to me thinks the market that is ~1/10 the size of the other is obviously a better place to work. I guess cost of living can have a big impact.
It's also multiple times the size of most states. Per capita is probably a better metric, where it ranks 34th.
> and economic output
I can't find the raw data, so I'm not sure what they're counting. It says "computer and electronics" is highest, which is 2x above chemical, which makes me very suspicious. I work in manufacturing that happens 100% in China. Would I be counted? Are they counting AMD or Qualcomm, where their chips are manufactured in Taiwan (or Arizona)? How much of this is military/government contracts (which aren't subject to many of these pressures)?
Tesla still seems to have quality problems, and never delivered the promised $30,000 car. Although BYD did.
> Although BYD did
I don't believe this is true, due to the < $30k cars not meeting US safety regulations.
I can't attribute their lack of presence in the US market to them when their cars face a 100% tariff.
That has very little to do with Fremont the facility.
The Fremont facility is where Toyota showed GM how to build cars properly--an expensive lesson which GM and Tesla both decided to ignore.
I would suggest that the real "innovation" of Tesla's foreign factories was simply getting them far enough away from Musk that he couldn't easily screw them up ...
Hopefully without all the illegal toxic waste dumping, forcing people to work during quarantine and ignoring crazy homophobic / racist environments and suing/harassing whistleblowers. The Fremont plant is not something to strive for in any way other than "it's a plant that manufactures things in the US".
> Mike Newcomb now leads the molding department. Newcomb obtained a business degree from a local college and now leads the molding department, overseeing production of the 4.3 billion pen barrels and caps the facility makes each year.
My question is always, why? Why was the degree required for this? He has 20 years of experience. Send him to a 4 week management training course or something. I went to business school for undergrad and it prepared me very little to run a molding department. I think this guys experience was already his biggest asset and this investment could have been significantly less expensive.
That said, as someone who doesn't have a business degree, shouldn't it ideally prepare him better for working with the rest of the organization? Dealing with budgets, strategic plans from upper management, marketing etc?
Management training would cover all this in a few weeks. He got a full undergrad degree, meaning half his courses were just core like history and English and all that stuff. It could have been much more focused is my point.
Part of the challenge of organizational management is making decisions structurally bound to prevent negotiations for every single possible response.
“Well Mike didn’t have a degree, so why not me?” Would become the next big thing
The “Simple” answer is, don’t have dumb gates for hiring and have your managers actually behave like leaders who have a personal stake in the success of their business and hires.
Other than rare, personality driven cases, this approach isn’t even available as a concept for people who get into business for money or power - primarily because “the market” doesn’t reward training people, because the last century has shown that people with money don’t want to take the risk
What evidence do you have to support this opinion as fact?
You?
>“Well Mike didn’t have a degree, so why not me?”
"Because you don't have the right degree" is an easier and less confrontational answer than the truth ("the position was filled nepotistically", "you know what you did but the incidents are not properly documented", etc.).
There's too much bureaucracy focused on CYA than being adults and properly talking with employees.
For example, the Shazam app. I knew right away it must be using Fourier analysis. But if one was self-taught, one might have never understood what the point of FA was, or even have been aware of its existence, and instead used kludgy, inept methods.
For a personal example, I was once given the job of taking the graphic display on a CRT and mapping it onto a printer page. The addressing was different, the axes were different, the pixels/per inch were different. I knew what the tool was - a transformation matrix. Had it ginned up in an hour and it worked first try.
A co-worker was completely baffled at this. He didn't know what a transformation matrix was, and likely would have otherwise spent a couple weeks on the problem and done a crappy job.
I.e. one doesn't know what one doesn't know. The advantage of an accredited degree program is the curriculum is selected by people who know what you need to know, and the order in which information is best presented.
For one example, learning accounting. A lot of people try to run a business without understanding how double-entry bookkeeping works. This gets them into trouble with the IRS, for one result, and with bankers, for another.
During summer vacation, at loose ends, I signed up at the local college for a course in double-entry bookkeeping. It's paid off for me ever since.
It is domain specific and sometimes even company specific. How such a thing can be standardised and formally taught is the question I'm interested in.
One of the reasons Andrei Alexandrescu and I worked well together as a team designing D is that he had the formal training, and I had the practical experience. Our abilities complemented each other's.
Independent invention is a stronger data point than peer review, or even independent reproduction.
Personally, I would not have bothered about their rudeness - although I know it's hard to totally ignore such comments.
I would be proud of myself, or at least be happy, for having invented the concept independently, more so if it was well known in computer science, because that indicates it had some utility.
The phenomenon is so common in tech--something I, too, have experienced despite computer science having been my major--that it should practically just be a right of passage for software engineers.
I do not believe such an issue is still a factor after the web and a billion education courses on computer science are available. Nowadays also with LLM's which will happily answer everything for you.
All of which implies that you need to know quite a bit about that topic (which you were asking the LLM about) in the first place, otherwise how can you distinguish right answers from wrong ones.
So lenkite's comment is not very useful or practical.
Education has its place. It remains the most efficient method of ingesting large masses of well-organized information.
On the other hand, people who learn from experts learn, in addition to the experts' wisdom, also the experts' biases and blind spots. Or, to say it differently, being taught by experts is a good way to learn how to do what the experts already know how to do, but not how to do what they don't. (And I say this without casting any shade on the idea of education by the experts—I am a professor by trade.)
The problem is not knowing what one needs to know.
The idea that you have to have some prior knowledge to recognize and solve a problem is bonkers to me. There may be some problems like that, but the vast majority of things have already been solved and therefore can be discovered by self learners being inquisitive when seeking solutions to their problems
I personally have narrow interest in a lot of subjects. This means I have a fairly wide vocabulary for digging deeper when I want to. And I have the internet as a shovel.
I've seen it in action plenty of times.
The trouble is not being aware of what you don't know, and wasting time going down an unproductive path. You may never realize there is a better solution. (You'll see this in programming where the self-taught programmer uses a bubble sort instead of, say, a qsort. Or using polling instead of interrupts. Or not being aware of hash tables. Or not being aware they shouldn't roll their own crypto.)
For example, I enjoy making things with tools. Sometimes I struggle with a tool. When I have a craftsman over to the house to do some task, I ask "mind if I watch?" They always say "sure", and I watch them do it. They're happy that someone finds their work interesting and love to explain what they're doing.
I keep track of the tools they use, and how they use them. Many times I didn't know those tools existed and they make the job a lot easier. I also will note down the manufacturer and the model of those tools, so I can acquire the professional version of a tool.
(Yes, I look at yootoob instructional videos. Watching it in person is far more instructive.)
BTW, the number of people who seem to be capable of self-teaching themselves calculus seems to be very small.
> wasting time going down an unproductive path
This is usually a fallacy. First, in most cases, what's the hurry? You're learning things along the journey. A self-taught programmer will encounter this problem once, then know about qsort for the future. No 4 year degree required. Second, if the bubble sort gave them the performance they needed, they wouldn't care to find the qsort solution. If the performance was a problem, it's likely they would stumble upon qsort as a potential solution. Third, we see professionals with fancy degrees and well funded companies build by smart people leak plaintext passwords all the time. Don't roll your own crypto is wise advice but I think anyone that searches "how to make a login system" or "how to store passwords" is going to find that advice. All to say, the 4 year degree doesn't necessarily protect you from making this mistake.
> Yes, I look at yootoob instructional videos. Watching it in person is far more instructive.
I have the same passion for making. I build homes as a hobby. Learned it from youtube. I find if I struggle with a tool, it's because I just haven't used it enough. However, I approach the same situation you mention a bit differently around professionals. Maybe due to my self-taught nature but, I do research on the problem ahead of time. I basically know exactly how to fix the problem but don't want to - or just lack confidence in my ability to execute it (drywall finishing is easy, in theory, I suck at it so I hire it out every time). Usually I look at the multiple ways the problem is usually solved and form my own opinion about how I want it done in my case. If you dig enough, you'll find the pros and cons of most solutions. There's usually videos comparing the methodologies on YT. I study all that as if I'm going to fix it myself from soup to nuts. So basically, I'm ask how the pro plans on doing his work, what products he's using, how long between steps (if something has to cure), what tools and material he plans on using, etc. Then, I will challenge his way if I think he's missing a step or taking a shortcut. I will also watch during his execution and tell him if something doesn't look right so he can fix it before it gets too far ahead. Of course, I ask questions and do so more friendly so as not to be a judgmental prick constantly critiquing them. But, I've already done so much research and seen the tools in use that I'm rarely surprised by how something is being used or it's existence.
Probably one exception is fine carpentry and furniture making. I never had interest in those things and there is a whole world of specialty tools that I've never stumbled upon. I also don't really have much need for that specialty as I just buy furniture and don't have a historical home myself, the modern architecture homes I build doesn't have much need for it beyond custom closet systems and cabinetry (which is built off site)
> BTW, the number of people who seem to be capable of self-teaching themselves calculus seems to be very small
Well, I'd argue that the number of people interested in doing this is already very small and the key is to be resourceful in your instruction. You can ask for help when self learning. But that doesn't mean you have to signup for a semester long course that covers one topic a week. Jump on Udemy or similar and I'm sure most people with the right drive, will learn it. "Self-teaching themselves" doesn't have to mean doing it yourself or re-inventing the wheel. However, when you don't need to learn something as part of a larger goal (like a pre-req for your 4 year degree) then it's easy to give up when it's a hard subject. People rarely need to know calculus unless they're in school.
But, people rarely care about actual knowledge, and want that degree checkbox checked.
That may be true for recruiters, but it is not true for the manager who needs a specific skill set.
Caltech has a large course catalog. I could only take a certain number at a time. Other than what was required for the degree, I carefully selected courses I anticipated would have the largest payoff in my desired career. The usual course was 3-0-6, meaning 3 hours of lecture a week, 0 hours lab, 6 hours study.
I don't regard any of it as wasted time and money. Except the economics class.
Here's my class load for one semester in my junior year:
AMA95 Intro Methods Applied Math 12
AM97 Analy Mech Deform Bodies 9
ME5 Design 9
ME19 Fluid Mech & Gas Dynamics 9
CS114 Micro Process Systems 9
BEM106 Business Econ Sem 9
That comes to an estimated 57 hours/week of work. For me, those estimates were pretty accurate. If you want to point out the waste of time and money, have at it!
I also don't use Fourier transforms much, or Taylor or McLaurin expansions - but it gives me a solid foundation for understanding how people do various signal analysis, function approximation, etc.
I.e. I agree that I had to take a lot of math. And I don't use much of it. But I'm not really resentful of it - I'm only resentful of the sometimes useless profs we got.
This seems to me like saying "I could be a better engineer by combining engineering expertise with mathematical knowledge, and I have that knowledge, but I prefer not to use it." I don't always learn new things well enough to realize their potential, but I've never once regretted learning something new, and, if I don't know how to use it, then I usually assume that's on me, not on the subject.
Education does have some value but I believe it is mostly for people looking to push the envelope and do research. For people who are going to be part of the industry/workforce it's extremely overrated, especially since all knowledge has a useful half-life that tends to be pretty short. If you pick people 20 years after their studies and ask them questions about the curriculum, they are probably going to be unable to answer much of it, outside of what they regularly use for their current or previous job.
Every liberal arts major could make this complaint. The curriculum is supposed to teach you to think, not necessarily to teach you things.
For people starting their career with no experience and no network, yes the undergrad signals everything you mentioned.
But the smell! Yuerk. I don't know what it is, but I find it awful. The smell is the only problem, but as far as I'm concerned it's enough that I'll never buy another.
The similar-quality European or Japanese permanent markers don't smell anywhere near as bad, but have surely lost market share here due to Sharpie's effective marketing.
The biggest problem is that the base underlying our manufacturing capability has almost completely evaporated. You need tool and die companies to build parts for your manufacturing machines, but almost none are left. There are only a few plastic injection manufacturing companies left, and they are either fully booked or on the path to shutting down. (No, you can't 3D print products for the mass market, it's way too slow and expensive.) Young people think of trades and manufacturing jobs as fall-backs that only drop-outs who couldn't hack college get into. You will make many times more money as a salescritter convincing people to buy things they don't need than using decades of engineering experience to build the things they do.
It's a popular misconception that China's manufacturing advantage is cheaper labor. This was _maybe_ true for a while, their advantage right now is that they have all the things mentioned above: they have the tool and die shops, they have the supply chains, and they have workforce. They can go from product idea to shipping in a week. I do not celebrate the Chinese government for much, but making sure they had a robust manufacturing base from top to bottom was the biggest and most important thing they have ever done. The US government let ours wither and die. Which has made our economy remarkably fragile and essentially decimated the whole idea of a blue-collar middle class.
Destin from Smarter Every Day encountered all of this when he tried to make a better grill brush using only a handful of parts and _still_ did not manage to make the whole thing using only US-originated parts. It's a fascinating window into how hard it is to make anything at all here: https://www.youtube.com/watch?v=3ZTGwcHQfLY
That's the fundamental problem. The pay for industrial roles is garbage compared to the skill required. Anyone talented enough to do those roles can obviously see that and opts to do something else.
Consequently, industrial employers don't have a local monopoly on jobs like they used to. Any industrial employer is competing against any remote position willing to pay above them. That generally isn't hard given how little the industrial employer is willing to pay.
2) Industrial locations generally suck and you have to be there in person
Manufacturing generally needs space, and real estate prices are ridiculous. Consequently, manufacturing tends to be located in undesirable areas. That means long commutes or locating in sub-tier cities.
3) No penalty for outsourcing to foreign countries.
This is what tariffs are supposed to be for. You slowly ratchet up tariffs until you get the result you need.
The problem is that tariffs get politically captured by the oligarchs who use them to completely protect their stupidity (see: the big automakers in the US in the 1970s) until they get totally obliterated.
And yes, they are enforced.
I've talked with some friends here in europe after seeing that video, and apparently tool and die remains more avaliable here. (Metal casting is impossible nationally though)
I suspect the larger travel distance to China from europe create a market nieche that keep it alive when quick iteration is important. I dont think we succeeded better politically basides that luck. (Trades is frowned opon in the same way)
Locally we have a good prototype industry for cirquitboard design (we work with them regularly), but very few large scale producers. The companies that make electronics nationally often have their own cirquitboard factories for their usecase.
But that doesn't work for new companies trying to create new products. The most common approach is to send the design to China and get a pallet of cirquitboards back fully assembled. (With frankly excellent expert support for any issues discovered in manufacturing)
That means any middle sized company could have this in-house if they wanted to.
As a mid-sized company, you care about IP protection. That means the price for China will need to include bilingual lawyers. Otherwise you might end up in the super awkward position that your mold gets copyrighted by the manufacturer inside China. That would then allow your contract manufacturer to prohibit other injection shops inside China from working with you.
Parts also vary in complexity; throwing a CAD guy without CNC knowledge in front of a 5 axis CNC and telling them to make a injection mold, doesn't feel too far off from giving a soldering iron to a software guy and asking them to design a circuit-board. It may work out if it's simple, but be prepared to invest a-lot of time letting them fail and build up a skill-set if you need something complex.
I have heard from an acquaintance that once you have a good mold, there is no reason to not just injection-mold nationally.
One way to help with this fight is to charge a tax on things made in Shenzhen. This helps by funding the government, helping build roads and fund things like healthcare, and also makes those items more price competitive with American-made goods.
Tarrifs can work. They're an excellent tool when applied stratecally, gradually, and combined with focused effort to nurture local alternatives.
Just because we're currently seeing them used in the most crude and reckless way possible does not diminish their value.
The comment should normally not have needed the context i added; to be taken seriously.
Or bet on china crashing their housing bubble and take their currency stability and industry with them. But we've been predicting that for well over 10 years and it hasn't happened yet.
... and it would be quite useful to have local industry to replace it with, or we just end up crashing the USD aswell when nobody can produce anything.
https://new.abb.com/news/detail/129685/abb-to-divest-robotic...
40 more comments available on Hacker News