Hacker News new | comments | show | ask | jobs | submit login
The Insourcing Boom (theatlantic.com)
173 points by airlocksoftware on Dec 4, 2012 | hide | past | web | favorite | 108 comments

American declinism has been fashionable this past decade or so, even before the economic issues. And there are still economic issues. But it has a history of being the first in the world to rise from the ashes of the old systems and build the future, and while past results are no guarantee of future successes, I am still optimistic that we are not yet spent.

American manufacturing will return. It may not be the job engine it once was, but it is still a wealth engine. American energy prospects are suddenly looking up, leading the world in both the tech to extract it and, this time, the resources to extract. We are only at the beginning of tearing down our scelerotic educational systems, but there is a good opportunity to build a truly 21st century education system over the next decade or two. We remain a high-tech economy. And we've got some wild cards up our sleeve, like a budding space industry and advanced medical research, to say nothing of whatever industry will come out of nowhere in the next ten to fifteen years that I'm not even thinking of to become huge.

I celebrate anybody who succeeds, and I don't particularly care if America is or remains "exceptional". I want everyone to succeed. I'm just saying that I think there's good reason to believe that for all the challenges in the world that America still has a very good chance to be in the set of successful countries in the medium term (5-25 years).

> American manufacturing will return.

It never left. The US manufacturing output steadily increased from the 50's to the recent recession. Right now, it is at an all time high. The idea of american manufacturing being shipped overseas is just false, and I'm sick of people repeating it.

What happened is that as wages got expensive, US labor was replaced by machines where possible, and by some cheap process steps overseas where not possible. Intel CPUs are a good example: They used to be made wholly in the US, employing a lot of blue-collar workers. To save costs, the factories were automated to the point where the only process steps left that employ a lot of people without advanced degrees are packaging and testing. And then those were shipped to Malaysia and Costa Rica. The high-capital process steps that create most of the value in the system are still in Oregon, Arizona and New Mexico.

Manufacturing jobs, as they used to be, are gone. Not because of globalization, but because of robots. And the old kind of manufacturing jobs will never return -- it just doesn't make sense to pay people $30 a hour for what could be done faster, cheaper and better by a robot.

What is the role of environmental regulations on locating silicon fabs?

A buddy of mine made a mint working as a welder building fabs in Phoenix when they were leaving California in search of cheaper water and lower regulations. He was very worried about the toxic emissions and very lax (non existent) oversight.

American Manufacturing is at an all-time high. It's not employing as many people, but that is due to efficiency, not decline.

Interesting article but I wish they had more examples than a heat pump based electric water heater and a $3100 refrigerator

The article is suggesting that jobs that can support a family are coming back, but with the described poverty-level $13.50 wages for skilled line workers that's obviously not the case. We're not talking high school dropouts either: among the tasks given to the production workers is redesigning from scratch products which use impractical and inefficient designs, at least according to the scenario explained in the article. The article also points out that these are lines where there are not many workers per line.

My own thoughts about the market were piqued by the description of the chosen product. Here we have an electric heat pump heater that costs over $1300. This is right at a time when running an electric heater, even an exceptionally efficient one, has become a poor choice given the price of gas. Furthermore, gas is very likely to remain cheaper than electric for many years. Not to mention that gas heaters are simply more efficient for heating to begin with. Rather than buy an $1600 chinese or $1300 american water heater, one can buy a $150 mexican factory built water heater that runs on gas and will save many thousands of dollars in energy costs over its lifetime. That's the rational choice, it is unlikely this complicated electric one is going to be something with a lot of market growth. Hopefully there are better examples of high growth reasonably priced mass market items that can be built as well, which have a chance at creating more than a handful of these $13.50 jobs (comparable to the salary at CostCo or McDonalds, BTW).

For water heating one could also do what I did - use $50 worth of materials and junk to build your own solar water heater that reduces the cost of water heating to almost nothing. (http://www.motherearthnews.com/renewable-energy/1979-09-01/A...)

We're essentially reaching an intersection between the cost of Chinese labor (steadily increasing as China's economy grows and their currency gains value) and US labor (depressed because we're still at 8% unemployment). If these workers still wanted to be paid $30/hour, those jobs would still be in China.

In December 2011 there was a similar article in the NY Times[0], which addressed the expansion of manufacturing jobs but had a much bigger focus on wages.

Unfortunately for those workers, wages won't go up until the US economy continues to improve and unemployment goes down. Until then that "just thankful to have a job" feeling will suppress labor wages in general. Last quarter, corporate profits reached a historic high even as wages reached their lowest-ever share of GDP[1].

Maybe I'm just raw from the hostile election season, but this has always been personally frustrating to me. Our private sector doesn't pay enough for the jobs it offers, yet any suggestion to have the government assist is considered "redistribution" or "socialism."

[0] http://www.nytimes.com/2011/12/30/business/us-manufacturing-...

[1] http://money.cnn.com/2012/12/03/news/economy/record-corporat...

and US labor (depressed because we're still at 8% unemployment)

Wages are not depressed. Per-worker wages have increased in both the public and private sector:


Sum[wages] / GDP is down because employers are finding ways of getting more output with fewer employees. I.e., Sum[wages] = (average wages) x (# of workers) is down because # of workers has decreased, not because wages have.

If wages had decreased, we would not have had a recession at all (according to Keynesian economics).

Eh... what is wrong with redistribution or socialism? I am all for them.

$13.50 isn't poverty level. At 40 hours a week, 50 weeks a year, it works out to $27,000/year, which is just about the median individual income nationwide.

This yields an after-tax net of $1,750 per month. You can rent a 2 BR apartment in Louisville for $450 per month. The $1,300 left over is enough to live quite comfortably in Louisville--heck I don't spend much more than that in New York City (the difference is I pay $1,500/month on a studio).

You just got out of college. You went to a good school, you got good grades, and you worked hard. There are no jobs in your field.

You're bringing in 1750 a month, but 500 of that is going to student loans. 450 goes to rent. 200 for heating in the winter. Add in the rest of your utilities, and you're looking at 1300 a month. Do you have a car? 300 a month. Do you want to eat? You can probably get by on 150 for food.

But you have no health insurance. You have no savings. You cannot start a family because you can't support a family.

And, unless you get very lucky, this is going to be your situation until your loans are paid off. You will be 35 when your adult life can actually start.

But if you're unlucky - if you get sick, if you need to take care of your family, if nearly anything bad happens, regardless of how minor, you're probably completely and utterly fucked to an extent that a lot of people can't possibly realize.

It you went to a good school, and got good grades, you probably are not working this job. It seems that it has become fashionable to believe that every job should pay upper middle class wages or not exist at all. Hindsight should make it completely obvious what happened when unions forced that issue. Those jobs went away and are not coming back, they were never sustainable.

These assembly line workers are probably not college graduates with $60,000 in student loan debt, nor are they people who can expect to have a stay at home spouse.

If you "just got out of college" then you are likely <26yo and are covered by your parents insurance.

A 2BR in Louisville for $450 puts you squarely in the bad part of town. Aside from it being less safe and generally less enjoyable to live in (probably noisy neighbors, people up all night, etc.), you also are going to be commuting 20 - 30 mins per day to the other side of the city where the decent jobs are.

But does that $1,300 support a family?

What family are we talking about? Just-married or with three kids, two cars and a dog, living in a rented apartment? There aren't that many things that can support irresponsibility or chronically poor decisions. Anyway, I'd say yes, it does, because a partner is bringing in his/her $1,750, too, and $1,300+$1,750 is pretty OK.

If your spouse also has a similar job, then you make $54,000, which is right around the median income in the US. You can definitely support a family on $2,500+ after-tax, afer-rent income every month.

I strongly agree with the overall direction of your comments on this subject, rayiner, and I think the workability of a median income in North America is vastly underrated.

... but it makes me a little sad that the median family household in this model requires both parents to be out of the house for 40 hours per week plus lunch hour plus commute. I mean, I think the Ward&June Cleaver myth was a little silly, and I don't think it's an Inalienable Right that every single family should be able to afford indefinitely supporting 4 people on one income.... but it seems like it would be better for society if kids got more parental energy than I can imagine them getting in this circumstance.

I mean, I'm not complaining about my job (ha, not at all), but I note that I'm barely a functional spouse when I get home after work, much less able to be a great father (not that I have kids). And even if I was, I'm not getting home until 7 or 8. Suppose my wife had the same schedule, what kind of (hypothetical) parents would we be?

I'm not saying it's Wrong or Unjust. I'm just saying it seems a little sad.

To be fair, these are 40 hour a week, 9-5 jobs. Very few $13.50/hour jobs require the kind of extended hours that professionals endure.

As for what kind of parents you'd be not getting home till 7-8... parents like my wife and I? The baby isn't old enough yet to really recognize either of us as anything more than a milk source, but when she gets older that will be her reality. I honestly don't think it's a big deal--my dad never got home till 8 each night, and traveled extensively and work the whole weekend, etc, and my brother and I have a great relationship with him. A lot of the "putting in face time" when it comes to kids is more for the parents than for the kids. Kids are perfectly happy playing with their friends at school/having the nanny take them around so long as they know that you're there and see you everyday.

>but with the described poverty-level $13.50 wages for skilled line workers that's obviously not the case.

A couple of points: 1. $13.50 * 2080 is $28,080, which is clearly not rockin' Prada but, in places with low costs-of-living, is very livable. Silicon Vally and places where HNers are likely to congregate are much more expensive, in large part due to housing restrictions (http://www.slate.com/articles/business/moneybox/2012/05/face...).

2. $13.50 is where workers start, not necessarily where they end.

3. I'm obviously not arguing that workers are making a LOT of money or that they're likely to get rich, but they're also not poverty-level in most of America and are still much higher than minimum wage and much better than nothing.

One ambulance trip can easily cost $8,000. Now you are down to $20,000 for the year - hope you didn't have any kids.

But the point isn't whether $28k is livable in Nebraska, but how it compares to what people used to get back in the day.

Honest question: does that still cost $8,000 under Obamacare?

What's Obamacare? My life hasn't changed.

And when Obamacare kicks in, my healthcare life will still be fundamentally managed by insurance companies, with somewhat changed regulation around the margins but still extremely profitable.

The margin on health insurance companies is like 8%.

You have to be careful there, because some of the largest consolidated insurance providers are nonprofits that are incentivized to maximize overhead (to the benefit of internal executives and managers). The ones that aren't have to compete with the ones that are.

Very true.

FWIW, I used to work on an ambulance and when the person didn't have insurance the cost was $500, not $8000. Not sure what we billed insurance though. Oh, and if they couldn't pay, we didn't follow-up. They were sent one bill and that was it.

I've always heard that was because medical debt was legally very difficult to collect, true?

So screw it.

I'm sorry but we can't constantly let the price of medicine / medical care drag down the world's economy as a whole.

"We've got this medecine that shall raise your life expectancy by one year if you have a cancer and it cost $20 000, let's offer everyone this".

"We've got this medecine that shall raise your life expectancy by two year if you have a cancer and it cost $100 000, let's offer everyone this".

"We've got this medecine that shall raise your life expectancy by three year if you have a cancer and it cost $100 000 000, let's offer everyone this".

"We've got this medecine that shall raise your life expectancy by four year if you have a cancer and it cost $100 000 000 000 000 000, let's offer everyone this".

Where do you put a limit of what's acceptable?

In the end it all comes down to death tables and one day politicians and everyday people will understand that.

If 99 persons can live decently with $28 K and the hundredth one can't pay the $8 K ambulance, so be it.

Because, in the end, it all comes down to death tables and we can't allow totally overpriced and unjustified crazy high medical (and related) cost, like a $8 K ambulance trip, ruin down the entire world's economy.

Also it would be nice if these arbitrarily high medical price ween't constantly used as a form of intellectual terrorism justifying more and more debt creation.

You absolutely can.

It's about returns to worker productivity. As the article mentions, factory workers are hundreds of times more productive than they were in the 1960s. They are probably thousands of times more productive than they were in the 1760s.

How much has worker productivity increased in, say, symphonic music? In the last 300 years, I would guess it has perhaps doubled, if that (amplification) And yet symphonic concerts still happen, despite costs increasing without bound!

Medical costs rise because new services are available, but worker productivity rises slowly. Medicine may be a larger and larger fraction of the economy, but as long as worker productivity rises in OTHER fields, we will still be able to afford it.

If only it were as simple as pill = dollar = life expectancy.

All I have to say is I'm happy to live in Canada and don't worry about these things.

Glad I don't have your taxes, though.

I love our Medicare system (Canadian) because it creates a peaceful feeling of reassurance. You show your Medicare card and you are good to go, no paperwork, no bills, nothing. It's hard to explain that feeling, the hospital is not a scary place, it is a safe place. It's there to help you. You just go to the doctor when you are sick, no big deal.

On average, we pay roughly 30-40% Income Tax. The less you make, the less you are taxed. If you make less than $38,000 in my province, you would pay something like 24% (15% federal tax + 9% provincial tax - http://www.cra-arc.gc.ca/tx/ndvdls/fq/txrts-eng.html). And if you make more you get taxed more, but it's reasonable.

Honestly, when you grow up here, you just get used to it. It comes off the paycheck just like retirement savings and employment insurance, and you don't think about it.

I'm a proud Canadian, equal healthcare for everyone is a value we share and I think it's worth paying for.

When I add together my taxes plus the health insurance premium that comes out of my paycheck, my net take home is probably less than what it would be in Canada. Sure I have the choice to not pay for health insurance, but that's an illusory choice. I've got a wife and a kid--I'm not going to opt out of health insurance.

Except that tax rates in canada are about the same as they are here in the states.

Sorry to go off on a tangent, but,

>Not to mention that gas heaters are simply more efficient for heating to begin with.

Are you sure about that? A heat pump with a COP of, say, 2 is 300% efficient (it is said to have an Energy Factor, EF, of 3). That is, for every Joule of electrical energy supplied to the pump, 3 Joules are added to the water tank. On the other hand, the combustion products of a natural gas heater need to be exhausted outside the building. Unless they are first cooled to the temperature outside, the heater is less than 100% efficient.

Where I live, a Joule from the electric company costs about 2.5 to 3 times as much as a Joule from the gas company. A quick Google search indicates that this type of water heater typically has an EF of 2 to 2.5. The cheapest gas heater I could find on the Home Depot website was over $300 (twice your Mexican one) with an EF of 0.62. So it wouldn't surprise me if the energy cost of an electric is on par with the energy cost of a gas heater, and it would surprise me if the electric actually used more energy. I don't see how you would save thousands of dollars in energy with a gas heater: Even in the worst case (a 2 EF electric heat pump vs. a 0.62 EF gas in a location with a 3x energy cost ratio), the gas only just breaks even.

Market effects affect to the economic perspective, but considering the transformation of fuel to heat, gas is sure to be more efficient. Electricity used to run the heater is generated by burning something and running a turbine - a process that itself runs with around 30% efficiency. Add to that the power loss from electricity distribution. 300% efficiency on the heater isn't enough to break even so that you'd get as much heat from the same amount of gas burned locally versus in a power plant.

Well, ok, so 30% thermal efficiency at the power station[1], and 10% loss in the transmission lines[2]? That gives us .3 * .9 * 2.5 = .68, or 68% thermal efficiency. That compares favorably to the manufacturer-claimed 62% efficiency of the gas heater I cited above. I can even imagine a scenario where an architect could design a house so that heat source of the water heater is near the heat sink of the fridge. In that case, the coefficient of performance of both devices would improve.

In case this wasn't obvious before, let me state this very clearly: Over 30% of the energy content of the natural gas used by your water heater goes up the chimney and is not used to heat the water. Gas water heaters and furnaces are NOT 100% efficient.

[1] Wikipedia claims that conventional external combustion (i.e., steam) power plants have a thermal efficiency of 33%.

[2] Wikipedia claims that transmission and distribution losses nationally averaged 6-7% in recent years.

This is neither here nor there, but it strikes me as interesting that the most efficient energy management in a home is just ensuring that your interior environment stays stable and isolated from the exterior as much as possible.

Okay, you got me. Still, burning stuff locally is in the same ballpark as running a heat pump with the electricity. Should microturbines become feasible to buy and operate in every home, it would be the most efficient way, running a heat pump by a turbine of your own and using the waste energy as heat source as well.

Maybe I've had too long a day and misread something, but here I go anyhow. The efficiency of a gas water heater may be increased through mechanisms that enable it only when water flows through it, such as when one takes a shower or fills the dishwasher. In contrast, a classic electrically heated water tank sits there to be filled with cold water, which gets heated up, and if it isn't all draining out already, cools back down to require reheating. Reheating a standing tank of water several times over the course of a day is wasteful. In a related thread, also consider the same water tank while one takes a long shower. We've all* (* This is a vast generalization for the internet) had the hot water run out of us during a nice shower, and this is because an electric water heater cannot effectively heat a freshly filled tank of water while said tank is being drained. I'm sure if another person feels the need to expand on this at all, they're quite capable.

There are plenty of on demand water heaters in both gas and electric.

Combination Boilers that provide both hot water for heating and hot tap water are advancing quickly. Flexible PEX tubing these cut out a ton of traditional cutting and brazing plumbing labor to install hot water loops. It's also a lot more efficient to move hot water to a heat exchanger of some form in each room than to push air through ductwork. Some of these rigs also can be easily combined with a solar thermal roof array, further cutting your gas or electric costs.

I'm pretty excited about the last option, since I have an older house with poor air return paths, which means my furnace is almost always ingesting chilly basement air rather than getting a nice recirculation of heating air that's already been heated. Not to mention radient heat exchangers are silent and allow easy individual room temperature control with the smarter thermostat systems.

I recently moved into a house with the worst heating system ever: In ceiling radiant heating. Apparently the house was designed without regards to the laws of thermodynamics. On the upside, we have probably one of the warmest attics in the winter.

If I turn the heat on in all the rooms, I can watch the electric meter spin.

If you have access to the attic and it's unfinished, check if there's a reflective layer above the coils. If not, that's easy to add and will make a big difference.

Not this is a _reflective_ layer, not just insulation. The thermodynamics of radiant heat systems involve photon transmission, not just gas convection. If it was installed by a general contractor they may not have understood this distinction and the need for aluminum couplers and metalized reflective insulation.

Thanks for that tip, I'll try it. I actually reasearched it a bit more, and see that it is possible to have in-ceiling radiant that isn't crap, it's just that mine is crap. Supposedly it responds faster than forced air, but mine takes about an hour before I feel its effects. Also one room has 12' ceilings, which is stupid.

I just moved in to a house that has a gas on-demand hot water heater and I basically can't figure out why we use so much energy to keep gallons of hot water on hand all the time. It just doesn't make any sense to me. I think the upfront cost was higher, but the savings across the entire system are more than worth it.

I think you're misinterpreting the article regarding production worker tasks. While they do suggest that production workers were involved - "By considering the workers who would have to put the water heater together—in fact, by having those workers right at the table, looking at the design as it was drawn" - they don't say that the production workers themselves designed any products, let alone from scratch. It's fairly clear that feedback from some production workers was taken into account during the design process, nothing more.

Mechanical design work of appliances is not something that you'd do alongside shift work. The workers getting paid $13.50 an hour are not going to be mech-eng degree holding designers.

There can be a huge difference in quality between goods produced in the US and those produced abroad, something that many people don't realize anymore because they've gotten so used to buying cheap foreign imports and replacing them more often.

When I was in school, I used to buy a new bag every year because they would just tear and fall apart within 12 months. Eventually, getting tired of this, I went online looking for a good bag. I found out about Tom Bihn[0]. Both their management and their production is done in Seattle. The prices are steep, but I decided to take a chance on them, given the good reviews on various sites.

I've now had that same Tom Bihn bag for five years, and it looks exactly like it did the day I got it - not a single tear. I don't think you could say that about any bags produced at an overseas sweatshop.

0: http://www.tombihn.com/

I was really impressed with Timbuk2, who make their custom bags in San Francisco. I bought one for a friend and while the design was fairly simple and the price was a bit high, it was solidly made.

Fast forward a few years, and I picked up one of their bags at a Mountain Equipment Co-op (not quite a chain store, like REI in the States). This bag also wasn't cheap, but when I checked the label it was made entirely in China. The quality is not bad, but I haven't had a chance to abuse it for a few years. I was mostly disappointed that they seem to promote the pro-American rhetoric while the majority of their channel goods seem to be produced abroad.

I'm impressed with those Tom Bihn bags, because they seem to be uncompromisingly made in America. I'm going to keep them in mind when I need a new bag.

Timbuk2 still manufactures made-to-order products in America, where just-in-time production makes sense. Who is going to wait six weeks for a custom bag?

Price-minimized, zero-margin, off-the-shelf crap for big box stores are a different story. Then they can trade on their brand equity built with embroidered logo bags and sell trash to the mass market.

OK, point by point:

- turn-around for custom goods from China is not 6-weeks. It took a solid 2 weeks to get the bag, which is what I would expect. The cost to ship expedited versus the labour cost is a wash, especially considering they stil had to post the bag to me anyways.

- The goods in-store are still marketed as premium goods; to use an HN appropriate analogy, this is like me telling you that your Mac is crap because you bought from a 3rd party retailer. For reference, this was $150 for a laptop messenger bag. Not a ton of money, but not cheap considering there's no leather, etc involved.

- If you look at the store I mentioned, they're not really big-box. They're in some awkward in-between yuppie place where everything is greenwashed. I still shop there, but I'm pretty disenchanted since this incident.

- This event has, in my mind, destroyed whatever brand recognition they had. The product is alright, but I'd rather support a company that does everything in NA.

The point being, the market're selling into would be more than willing to absorb the cost of local production. I suspect at a point during their growth they were unable to afford to scale up locally, and decided to contract volume manufacturing for the channel. Now that they're well established, I would hope that they would perform an exercise like the one in this article, and see if moving production back makes sense.

Unless you have it done by air, it's about five weeks to get things in from China -- four weeks of shipping, one week of customs.

(This is actually mentioned in the article)

I've had small parcels shipped by air from China (eBay, mostly). They typically arrive in less than a week. They could drop ship directly from China to the customer.

My point was, the cost of shipping by air versus hiring an American worker is probably a wash. I suspect they do their custom work in SF because it's a token gesture. Sort of local-washing, versus green-washing.

It's neither here nor there, but I'll happily vouch for the quality of MEC's backpacks, I've used them throughout a sizable chunk of my life, and find they require replacement owing more to folk with sticky fingers than wear and tear - and I'm not the kindest to my backpacks either!

I'm conflicted about just moving to MEC's house brand; I always hear awesome stories about how they source them to be indestructible, and treat their customers very well, but the designs are pretty basic, and they're a dime a dozen in Ottawa. I feel like I'd be at risk of accidentally swapping bags every time I set it down.

Timbuk2 Bags are amazing quality.

As long as we're pimping good American brands, I want to give a shout out to Allen Edmonds, which makes all of its dress shoes in Minnesota. About 3x the price of the cheap crap you can buy at the department store, but they'll last you forever and can be refinished (with new soles, etc) for a modest fee.

Not all. Just saw a pair that was marked as made in Dominican Republic. Your points about quality still stand though.

Yeah, for some of the models (I think loafers and more casual shoes?) the leather uppers are hand-stiched in the Dominican Republic, ostensibly because they can't find people to do the job in Maine where they used to do it. Most of their shoes are completely made in the USA, though, and I hope it stays that way. Otherwise I might have to switch to buying Aldens...

Funny you say that, I've got a Tom Bihn bag lying around as well with stitching and everything still intact as well. I've probably had mine for around the same time, I've carried everything in it from stacks of books to laptops & heavy bottles of water and not one single frayed stitch or piece of material.

I have a Tom Bihn bag too, but I don't know what kind of bags you were buying. I had an REI-branded backpack from age 13 or so until I was about 23. It broke, and I took back to REI (they'll take anything back) and got credit towards a new backpack, which is still going strong.

Another user of Tom Bihn bags. I was teased a bit for paying so much for laptop bag... But also after five+ years of use, it's still good condition.

I don't think it matters where a product is built. It's the design and build quality that makes the product a good value. Initial cost was high but made up by not having to buy another bag, not having to deal with a torn bag, and not having to buy another bag makes it worth it.

I found the story they give about GE forgetting how to make water heaters interesting. The line you usually hear about startups creating their core technologies in house and outsourcing non core things comes to mind.

Was GE originally known for manufacturing (or ability to manufacture) and then lost that core ability? If they did, what would their core ability be after outsourcing, management of manufacturing? Maybe I'm not making a good connection there, but I think it could be.

I find dark humor in that we've lost such crucial skills. In the US, we've nearly lost the technology to create water heaters, but can definitely make fart apps.

On a lighter note, maybe this will reintroduce some needed humility to American culture. It's easy to forget that each item we touch, everything we use, was put together by another human's hands. Soldering together the proto shield for my Arduino deeply touched me, reminding me that every resistor and LED and processor in my laptop was soldered in by another human.

I may be romanticizing it, or maybe I'm just getting old, but I'm starting to appreciate objects crafted by humans more than those generated by machines.

Actually it's likely that most(all) of the resistors and LEDs are SMDs soldered using reflow or wave soldering.

Your electronics from before the 80s (and often the 90s) was soldered by a human though.

These days just about any surface mount part should be assumed to have been attached by a machine, and for very good reason.

I am capable of assembly with 402 and 603, but I can't even begin to imagine that going well on an assembly line.

(805 is somewhere around the size of small rice, 402 is half that size)

Yeah, I don't mind doing QFP and SMD resistors (actually QFP is pretty quick when you get used to it), but individual resistors take forever.

Also, I forgot to mention most processors are BGA packages AFAIK, so impossible to actually solder manually.

Regarding GE, you can eyeball their public financial statements for a rough idea of their manufacturing:financial ratio:


They conveniently split their balance sheet between GE Capital and GE (industrial/manufacturing). Roughly 70% revenue and profit from non-financial (mainly manufacturing - from 747 engines, to trains, to energy, to MRI machines, etc), ~30% from GE Capital.

So yeah, always have been and still very much a manufacturing company.

As a more general answer to your question, in case you missed it, there was a good article and discussion a while back on companies losing manufacturing through outsourcing, and its consequences:



TLDR: “So the decline of manufacturing in a region sets off a chain reaction. Once manufacturing is outsourced, process-engineering expertise can’t be maintained, since it depends on daily interactions with manufacturing. Without process-engineering capabilities, companies find it increasingly difficult to conduct advanced research on next-generation process technologies. Without the ability to develop such new processes, they find they can no longer develop new products. In the long term, then, an economy that lacks an infrastructure for advanced process engineering and manufacturing will lose its ability to innovate.”

An interesting counterexample that demonstrates the point is Intel - they've maintained their manufacturing capability, and as a result lead the world in lithography and process technology, a competitive advantage that allowed them to compete with AMD in the mid-to-late 2000s even when AMD's chip designs were better, and to dominate AMD now that both Intel's process technology and chip designs are better.

Andy Grove has apparently dedicated his retirement to advocating for reshoring manufacturing for the deep competitive advantage it confers [1].

[1]: http://duckduckgo.com/?q=andy+grove+manufacturing

FWIIW, GE Capital was closer to 50% of the revenue before the recession kicked them in the teeth. Hard. Indications and rumors had it that GE Capital nearly took the company down in the 2008/2009 time frame.


"Loophole Helps GE Benefit From Bank Rescue Program" (article from 2009) http://www.washingtonpost.com/wp-dyn/content/article/2009/06...

Warren Buffet did a $3 billion "private bailout" in October, 2008 http://www.bloomberg.com/news/2012-09-14/buffett-ge-bet-pays...

GE has essentially been a mix of holding company and leveraged buyout firm for a long time now. I think of GE as a stock trader who takes really, really long positions in an industry as a whole by acquiring a subsidiary in the industry rather than buying the corresponding index fund.

GE is/was one of the best manufacturing firms in the US. The "Lean" in "Lean Startup" comes from "Lean Manufacturing". GE probably did more to innovate in Lean Manufacturing (and Six Sigma) than any other company in the US.

Product design, sales and marketing. Would have to read some more to see which area was really their competitor advantage in the 60's.


Simply put, if you manufacture in-house as well as design, both processes innovate each other. But if you outsource manufacture, it's a black hole; very little flows the other way because of language and cultural barriers.

That assumes there is enough juice remaining in manufacturing optimization. After a certain stage in a product's lifetime, you've done the bulk of the optimization and doing manufacturing in-house could bring minimal benefit.

Let's not swap one assumption for another perhaps hidden one. In this case, the assumption that the product life cycle is long enough that you ever have "mature" products. The article points out that the appliance product life cycles used to be 7 years or so and now they're 2-3 years and that a good bit of the monetary advantage to building in the US came from product improvements that arose from co-locating engineering and manufacturing and getting them to cooperate.

I thought this was an interesting article, if a little light on any hard numbers outside of one example (G.E.). Another anecdote that supports this is Apple manufacturing a limited number of their new iMacs in the U.S. http://9to5mac.com/2012/12/02/is-there-some-secret-imac-asse...

I'm a little surprised they didn't mention advances in automation / robotics that make cost of labor even less of a reason to offshore. The U.S. manufacturing sector produces more than it ever has, despite the fact that manufacturing jobs are down from 19.4 million jobs in 1979 to 11.5 million jobs in 2010. http://archive.mises.org/17964/u-s-manufacturing-output/

Unfortunately confusing manufacturing output with manufacturing employment seems to be somewhat persistent in the US. The employment numbers going down are seen as a bad thing, whereas it is the total output and productivity that matter. If the entirety of US manufacturing could be done by one person than that would be great!

We've been through this before. In 1870 US agricultural employment was around 75% of the whole workforce (~29 million people). Nowadays it is closer to 2% (~3 million people). We produce orders of magnitude more food with considerably fewer people.

It wouldn't be that great, because no-one would be able to afford all the goods that were being manufactured...

This is an actual first-world problem, and dealing with it is going to be one of the most important and interesting challenges of the next 50 - 100 years. I see technology continuing to improve productivity, drive down demand for labor, and exacerbate current inequality issues.

Our economic system isn't really setup to work well in a situation where supply of labor drastically exceeds the demand for it.

There are many academic pieces and work on the topic such as http://en.wikipedia.org/wiki/The_End_of_Work

Obviously some of the difference will be soaked up by new professions that didn't exist before. For example here is a list of jobs that didn't exist 10 years ago http://finance.yahoo.com/news/10-jobs-didn-t-exist-175608243...

I am convinced that prosperity is being driven by trade and specialisation as put forward by Matt Ridley http://www.ted.com/talks/matt_ridley_when_ideas_have_sex.htm...

Specialisation is a form of skill and so the issue is really what happens to unskilled labour. Everyone "knows" that education is a fix, but current education systems are extremely broken and closely modelled on the dawn of the industrial age https://www.youtube.com/watch?v=zDZFcDGpL4U

It will be interesting to see if the cost of living goes down since that will also greatly alleviate things. Increased automation should result in that.

Unless you start out with piles of capital, you have to get paid for doing something in order to pay for things like food. If your labor is worth almost nothing, you will be doing a lot of it in order to cover basic needs. That is only prosperity for people who have the capital to take advantage of low labor costs, for others it is hand to mouth.

There isn't any reason to suppose that demand for engineers and managers will ever come anywhere close to 100% of the working population.

Meanwhile, the advantage of theft, violent crime and the black market improves dramatically as wages decline to nothing.

You are assuming capitalism for everyone which is not a given. Heck it isn't even how things are at the moment. It is increasingly possible to shift the proportion of people paid for labour/capitalism to being supported by the state. If the costs of living go down (a likely consequence of increased automation) then the financial burden would decrease too.

It's not a first-world problem. Many countries have already solved it. It's very much a US-centric problem.

This is a frequent but incorrect and dangerous claim because it assumes that a capitalistic economic system is necessary and immutable when in fact that is not the case.

Incorrect because if productivity were increased to the extent where labor costs were no longer a part of the equation, then humanity would not cease to exist as this claim implies (i.e. no jobs begets no money begets no food begets mass starvation begets extinction).

Dangerous because this line of thinking drives protectionism (i.e. unions, trade barriers and tariffs, subsidies) which increases the cost of goods which provably does lead to homelessness and starvation. This is not to say that tariffs and such are intrinsically bad because there are other factors involved (e.g. national security), or that increases are not eventually offset by corresponding decreases as competitors enter the (free? maybe...) market. Nonetheless, people suffer until the eventually happens.

On one hand, outsourcing rapidly transferred several decades of manufacturing and design knowledge to GE's soon-to-be competitors. (I haven't seen a Haier fridge on sale in the US yet but it's a matter of years, not decades.) On the other, maybe it juiced executive bonuses for a few years and may have permanently removed organized labor from the US private sector.

I wonder to what extent insourcing represents surrender for US firms seeking access to the Chinese market. China's leaders don't seem keen on idea of letting foreign firms sell in China longer than it takes for Chinese firms to pick up knowledge. The licensed production of Russian Sukhoi-27 warplanes comes to mind - the Chinese agreed to pay for 200, assembled from kits with Russian assistance. After 100 had been built, the Chinese said "no more" - and started advertising an indigenous copy for international sale. [1] http://en.wikipedia.org/wiki/Shenyang_J-11#J-11

I have seen Haier products, including a small fridge, in stores since about 2 years ago.

All kinds of products from Haier are available on Amazon. For the appliances, it's one of the three brands that qualify for Amazon's professional delivery and installation service.

This was a really interesting article. It's interesting the misconceptions that some people have about outsourcing and the perception that it's cheaper. When you factor in transport, the price of oil and rising cost of living everywhere in the world (especially China) it makes sense that less reliance on petroleum and outsourced labour actually works out better in the end. It was once standard for products to say Made in the U.S.A and then the outsourcing boom kicked in and people got used to Made in China for electronics or Made in Indonesia for a lot of shoe-wear. Buying a product made in your own country is currently considered a premium product purchase. For example buying a hand-wired guitar amplifier or even hand-made guitar that is built in the U.S commands a much higher price and some are happy paying it because the quality is considerably higher, attention to detail and intricacies like soldering and component placement.

I would really love to see the manufacturing world return to the mantra of, "this needs to be designed and built to last as long as possible", it won't happen but it would be nice to live in a world again where the best components and design decisions are made with the consumer in mind, not lining the pockets of CEO's. Once upon a time companies created products that lasted forever, I've seen televisions from the 50's with original picture tubes and components still working.

It's good to hear General Electric (a company I am not overly fond of) have started manufacturing products back in the United States again. Companies like GE returning jobs and manufacturing back home are the only hope this economy has.

there never was a time when everything was built to last.

sure, to this day you can find old stuff that is still working perfectly. but what you don't see is all the other stuff that has been discarded, thrown away, replaced. survivor bias at its best.

right now products need to adhere to safety standards, energy efficiency standards and have green certificates. if something breaks and just puts a minor scratch into you major lawsuits are going off.

in the past, you simply died in an accident (whoops, sorry about the electrocution or yes, our cars sometimes go up in flames).

Here is something that's easily overlooked: bringing design and manufacturing together promotes innovation. Having a more rapid iteration cycle is important, but why? Here's a principal reason: it allows design flaws to become visible in the manufacturing, and leads to a redesign if necessary.

Having both these modules in close (spatial and temporal) proximity helps minimizing efforts spent in irrelevant problems. The general rule holds for most softwares with 'manufacturing' replaced by user appreciation.

The hidden cost of taking manufacturing offshore is that a GREAT deal of industrial knowledge is lost. How do you know a problem needs to be taken into consideration if... you don't even know what the problem is?

The glorious days of Bell Labs seems like heaven for tinkerers. Designing, implementing and manufacturing camps don't alienate but simply part of one living organism.

Here's a complex sitting empty in Detroit for over fifty seven years that dwarfs the GE appliance complex,the old Packard plant.


Just finished reading this. Overall I agree with the main point (that quality control is easier to deal with when onshoring versus offshoring) but I think that that's something China can account for and improve on in the relative short term.

The thing that China CAN'T overcome at least in the short term is both the cost of shipping and the amount of time shipping takes. As the article mentions, a shipment of product can take up to 6 weeks whereas products produced in the US can be shipped within hours of completing production. There's just not any reasonable way for offshore production companies to overcome this issue that I can see in the short term and that could drive the onshoring movement aggressively, especially with the constantly increasing costs of fuel.

I see this is a problem for smaller companies just getting into outsourcing and for brand new products. There have been examples on here like wakemate that have had a really hard time getting quality products produced in a reasonable timeframe.

For bigger companies though with more stable product lines I think this is much less of an issue. Bringing manufacturing back is working for GE for more complex and expensive products. I am not sure you will see then bringing back the basic models of appliances though.

It sounds weird, but I would encourage small companies to look at manufacturing locally. A company I used to work for in Ottawa manufactured small volumes of cameras locally (that really narrows things down). They used other companies in the area to etch and solder the boards (automatically), and then final assembly was performed entirely in the same office building where I worked.

The best part was, everyone was involved in the whole 'stack'. The engineering group I worked with handled QA, but also travelled downstairs regularly to help with manufacturing procedures. We watched assemblers and proposed design changes to make their jobs easier. The process created a virtuous cycle where the initial person laying out the board knew what kind of headers would be hard to connect in which enclosure (for example), and as a result the quality was very high while being competitive on cost.

I think an assembly process like that would have helped, for example, with the teething problems experienced by the Jawbone Up.

Probably not, though if the trend in fuel and energy prices continues that may come to dominate the equation even for those things. The other thing is working capital - if it takes 6 weeks to ship and clear customs you will have a lot of capital tied up and that has an additional cost as well.

True, but that is also a competitive advantage to these large companies. They can afford to have the capital tied up where smaller competitors can't.

I have been actually pondering this for software for some time now. I used to manage a medium sized development team that was concurrently distributed between the Virginia, Argentina, Ukraine and Pakistan. I've always thought that the amount of communication overhead due to language/layers and time lag caused by that actually increased the cost of the product.

For example, during the day, i'd get requirements hand it out to the leads. If the lead is at PK, then guess what, i'm getting up at 3am to talk to them. Sending emails wont cut it. He then has to translate to a different language and hands it out to his devs. Thats several degrees of separation between the analyst and the dev. I've always believed that what was accomplished could have taken half that team if they were all local.

I think software development insourcing is ripe for this in the U.S. There is a large cost of living divide between NY, SF and Missoula, MT and the mid lvl salaries i've seen here would go a LOT further in those parts of the country and the time zone and language issues are minimal.

They are also applying this to rebuilding their local IT: http://gigaom.com/2012/11/29/ge-needs-the-data-analytics-min...

Immelt owns NBC and is on the White House board of economic advisors, so has the political juice to get waivers and ward off regulation. If you are a regulator, you will pick an easier target, like Exxon, which doesn't own a news network. Even granting the things in the article (tighter feedback loop, higher gas prices for shipping, increased labor productivity), regulation is not mentioned, so the generalization to other manufacturing companies is suspect.

Immelt owns NBC

--part of *

Manufacturing is converting from blue collar to white collar.

It's more important to be a process developer and a technologist to operate the machinery and design the assembly process than it is to actually turn a wrench.

I can only imagine 3d printing and accelerating product cycles will increase this effect. Dev ops in the real world - the designers and the manufacturers become the same people and the result is 'autonomation' to use the toyota term.

So what kind of tax benefits did GE get for moving production to Kentucky?

It's interesting to consider the change to insourcing manufacturing in light of the more recent trend of outsourcing (or, probably more accurately, globalizing) white-collar jobs. Although you wouldn't get the same benefits in terms of saving shipping costs, the other benefits of the article seem like they still would apply.

At a large corporation I worked at a few years ago, there was an ongoing push to offshore jobs to an internal group in the same company. We were encouraged to use this group by our chain of command despite the fact that previous efforts at outsourcing in our own group had failed disastrously. So it would be good to see the public discourse swing back toward the benefits of insourcing so that managers have more ammunition to fight for what works.

There was a passing mention of union dynamics in all this. Without outsourcing, a lot of the improvements could never have been brought on board.

Redesigning the product, reducing effort needed to build it is all good , but eventually this will go offshore one way or another

The only way is constant innovation and engineering , but if people don't go for STEM education , only lowly paid assembly job and sales job would be left here

Well written article - it grabs your attention with a curious story and doesn't let go. In my experience quality of articles in The Atlantic is up there with The New York Times or The Economist.

I wonder if this is more a result of exceptionally high unemployment rates making locally-sourced assembly line production practical once again...

Applications are open for YC Summer 2018

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