One of the things that I like about your blog is that you are critical of economic data. So many people go to extremes these days either worshiping the chart or rejecting all statistics. Maintaining a judiciously critical attitude is the difficult position.
Another thought that popped into my head while reading this post is the fact that military effectiveness demands both constant improvements in equipment and human capital. Many are saying that we have over invested in gigantic ticket items, such as fighter planes, aircraft carriers, and tanks--while under investing in low ticket items, such as drones. What we must absolutely be best at is coordinating material and human beings so that we can find out where to attack and concentrate our forces on it.
I think that often the back seat reformers underestimate the capabilities of those big ticket items. After all, sufficient investment in EW capabilities may make those drones completely innefectual in a few years. People are often wrong about which machines will be useful in the future. The Bradley was ridiculed mercilessly yet it seems to be one of the most useful vehicles we have sent to Ukraine.
Pretty much all MBT's like the Abrams and Leopard II are considered more useful as long-range snipers, doing shoot-and-scoot. Bradleys are lighter, more agile, and more numerous; a combo of reasons they show up more often on the forward envelope of engagements.
Thank you for all that useful information, but it doesn't change the fact that the Ukrainians withdrew the Abrams from the frontline because of their vulnerability.
We sent 31 Abrams, 5 of which have already been lost; while we've sent 252 Bradley's. Ukraine has also lost most of its 90 Leopard II MBT's in combat, as well as some of its Challengers. My *guess* is the surviving MBT's are being kept in rapid reserve units outside of drone range.
The survivability of MBTs is evolving so rapidly that it is impossible to know where we will be in 2025. But since they were designed to be used in warfare of maneuver, it's no surprise that they are not much use in positional warfare.
The most important aspect of military effectiveness is having the social cohesion for sustaining a war long enough to win.
Afghanistan proved this can be done (even if the goals were wrong).
On the other hand, Ukraine and Israel are showing increasing politization/polarization of "defense of democracy". It seems only wars started by the right will receive public support. The right's isolationism is a military problem.
Afghanistan was a prime example of an unwinnable war. It was an unsustainable $90bn/year GOP-created money pit, that collapsed practically to the Kabul city limits only a few months after the Trump Admin withdrew air support & logistics to the Afghan Army.
The notion that Afghanistan was "winnable" is entirely due to the bullshit statistics the DOD released in an endless brown torrent of slime through every presidential administration after 9/11, where "winning" existed only on paper.
I'd like to see a deeper dive into where this data comes from actually. It's good that noah points out the possibility of data problems but articles about GDP or productivity trends are ten a penny and always seem to gloss over where the data comes from.
Love the blog, but I think you're overlooking Learning Curves (Wrights Law) as the primary factor behind declining manufacturing productivity. These curves state that production costs drop significantly with each doubling of units produced, but only for standardized items made in factories.
This explains why housing and solar costs remain high (not factory-made), Moore's Law exists (it's just Wrights Law for chips), and market share matters (more volume = lower costs).
Most U.S. manufacturing has shifted from high-volume factory production (where Learning Curves thrive) to construction, repair, and custom work—none of which benefit from these curves.
Case in point: Dayton, Ohio. In 2007, the city have over 400 machine shops, many producing standardized parts for giants like GE, NCR (National Cash Register), and Delphi Auto. However, Delphi left in 2005, GE in 2008, and NCR in 2009. Today, less than 100 shops remain, primarily focused on repair work, custom projects, and small-batch manufacturing—work that doesn't benefit from learning curves. If you were to analyze Dayton's manufacturing productivity today, it would likely show a steep decline, but this is because the nature of the work has changed. Volumes matter.
Learning curves are also the reason I didn't go bankrupt. In 2016, I bought a struggling equipment manufacturer. When a supplier threatened to bankrupt me, I tried making their product myself. My costs were 6x higher than theirs. However, a 15% learning curve predicted I'd break even at 80 units and match their costs at 2,000. I went all-in, and within 3 years, we were more profitable than ever. No amount of money or technology could have sped up that process; we simply had to produce repeatedly until we became efficient.
Learning Curves are real, but they only apply to specific types of production. The U.S. has largely moved away from those types, which is the real reason behind the decline.
If you want more info, there are good sources here:
Housing and *construction* costs remain high because it’s not factory made. Solar is taking over the world :)
Learning curves are also why solar is going to beat wind because more of its BOM is in the factory, wind has higher transportation and construction which doesn’t benefit the same.
As an entrepreneur who started two high tech manufacturing companies, I agree that the availability of capital is challenging for manufacturing capital goods. We did it, but we were the 'odd duck' in that we used capital for manufacturing assets. I agree that the capital markets in the US are not oriented towards such investments, but could be encouraged by government policy like tax incentives for manufacturing investments.
I wonder if you combine the limited capital available for manufacturing and the use of profits for stock buyback, then you see a big effect. Hard to get big manufacturing productivity gains without easy capital access. Look at how easy it has been for quite a while to get capital for various software ventures (the Silicon Valley VCs, also the current AI hype).
Now, it is possible that the lack of capital is because sources don't trust the profitability of manufacturing, but also quite likely that it doesn't show the ROI immediately enough to satisfy today's very short term stock market thinking (and IPOs)
One way to look at it is that plenty of money is also invested in bonds, which does not provide a return above the interest rate (unless purchased distressed). Capital goods should be financed from loans or other debt-like structures, with the main concern here by investors is the security of the principle. Here is where government incentives or alternate structures could help. A key aspect is pooling loans across a portfolio. Then tranches can be separated, analogous to how this is done in mortgages. this may be a bad idea: maybe a Freddie or Fannie for manufacturing investments? Another way this could work is for government absorbing first losses from the portfolio, thereby reducing risk. In any case, the government has figured out how to do some version of this, as seen by the success of the TARP program and the energy loan programs. So, there is a pony in there somewhere.
The electrons vs. atoms duality also biases for electrons (tech) because atoms (manufacturing of material things) are asset-burdened enterprises.
Theoretically, a tech company can maneuver faster because it doesn't have to buy or sell property, plant or equipment. Everything a manufacturer does has to be done slowly, even for a declining unit. It can take several quarters to go from the decision to close a plant to actual shutdown (disposition of labor, finding a buyer or salvage value for labor, selling the property, possibly being required to remediate the property in order to be sellable, etc.). The investors feel, "If I can move -illions of dollars with the click of a mouse, so should the companies I invest in."
Even if we continue to tax business income -- which we should not -- at least we should allow expensing of investment. [This would also have the advantage of eliminating the distortion of faster depreciation for some sectors and assets and undermine investment tax credits for certain activities.]
Why eliminate the distortions? They lead to investing more or less in the favored/disfavored activities rather than in those that maximize income. If youmean tht the incentive effects ae good, then it's probably because it is good to have more OUTPUT of the favored sector, not more INVESTMENT, so the incentive should be directed at the output.
From an economic point of view you may be right, but countries are thirsty for money. The trend is for all companies to pay at least 15% corporate tax, including those based in what used to be called tax havens.
If all corporations were structured as LLCs, then that might fly. It just seems reasonable that, for example, a US business benefits tremendously from the strength of the US military and the resulting stability in the world, and somehow businesses should help pay for that privilege in due course, as every other person/entity does as well. If that were via a flow-through entity and paid through personal income taxes, I can see that (though I would really hate paying tax on non-cash income). In the end, as imperfect as the government is, we (Americans) derive tremendous benefits from its existence. It does need to be funded, and those in charge need to be held accountable. Very imperfect, but what a country we have the luck and privilege to live in.
Stock buybacks are essentially a return of cash (capital) to stockholders, analogous to dividends but obviously quite different in effect. They are also a way of funding generous stock option incentives for executives (a way to juice compensation), and thus there is a personal interest by managers in making this choice in use of capital. That being said, returning cash to shareholders is an indication that a manager has no greater productive use of that cash. By increasing incentives for investment in the business, it would increase the amount of capital used for that purpose. However, discouraging buybacks would best be done through other means.
You mentioned overregulation of land use, and I think it’s highly probable that it hurts industry productivity specifically because it reduces “circulation of workers between manufacturing companies.”
Industry in the US has been spreading out, especially out of the northeast and Midwest and into the sunbelt, and its logical to expect that industries that scatter themselves among accommodating sunbelt towns rather than concentrating in industrial cities will have less circulation of workers.
Is this really true that American industry wasn't spread out before?
All I really know about is what I read in the book Janesville and I got the impression that it was pretty common for US manufacturing in early- and mid-century to be scattered among accommodating midwest towns rather than concentrating in industrial cities.
I mean, that's where the whole concept of "The Rustbelt" comes from, isn't it? That small cities like Flint and Janesville used to have high paying manufacturing jobs despite having populations under 100,000.
A quick Google finds this article[1] which claims "During the 20th century, America built thousands of manufacturing plants in small cities in the Midwest" in places like Danville, Newton, and Youngwood.
The movement to the sunbelt could also reduce labor productivity. Companies have been moving manufacturing jobs from high-wage union states in the North to low-wage non-union states in the South. It makes financial sense to move South even if your new workers in those states are less productive.
that may also lead companies to substitute capital for labor. It may be worth investing in industrial robots to make high-wage workers more productive but not worth the investment in lower wage areas.
But doesn't (modern) manufacturing need increasingly educated/trained workers? The sorts of employees produced from high-end Polytechs found in larger cities? I mean virtually all machining, for example, now seems to be computerized. The days where it was all done by hand by goes with only a HS diploma is long gone.
High wages in US tech/finance sector is really crowding out manufacturing
Is there any possible way to construct a two-speed America, like some envisioned for Europe about 20 years ago? Maybe some states can voluntarily give up US dollar so that they can do manufacturing on the cheap?
I don't necessarily disagree, but at the same time doesn't that suggest a surprisingly small pool of talent? Does that mean that even a country as large as the US can only be outstanding in a handful of areas? To be fair, I think it goes beyond tech and finance to also encompass life sciences, maybe healthcare, and maybe media, although I'm not sure how much media jobs would compete with manufacturing ones for talent.
Another hypothesis is there has been falling demand for US manufacturing. This has been driven in large part by the increased competitiveness of manufacturing first in Japan, then Korea, Taiwan and of course China.
Less demand for US manufactured goods made investments into new machine tools and equipment less compelling. With the benefit of hindsight, I'm not sure how the US could have kept many manufacturing industries alive here. However, we almost certainly could have done more to incentive more advanced manufacturing to stay.
For example, what would it have taken for Apple to produce iterations of the iPod and iPhone here? The iPhone is an extremely underrated reason for China's supply chain strength in consumer electronics that has spilled over in everything from EVs to drones and now eVTOLs.
From my read you looked at the customer side of the investment - we also have issues getting funding on the supply side. Automation will require a lot more breakthroughs to become economical.
An article on that subject in the NYT, years ago, claimed that Chinese industry can hire hundreds of engineers overnight. In the USA companies can’t hire _one_ engineer overnight. The background check takes two weeks.
Might be valid, but hard to see how that can have a major long term effect - the engineer still gets hired. Just makes China more nimble. To have a long term effect, we would need to assume that the slight nimbleness creates a big advantage. Not defending the slow hiring, just don't think it makes a big impact.
Add the effect of payroll taxes, and especially payroll tax accounting. The paperwork for a dishwasher who works for a week is almost as expensive as for a skilled technician who works the whole year.
You must be thinking of Ex-Im Bank credits. Sure that kind of credit can promote exports. But the macroeconomic machoism is that a deficit means, mainly, someone bought a USG bond instead of a private sector bond and the private sector bond was more likely to finance investment than consumption. So deficit transfers resources from investment to the expenditure. A tax reduces to some greater extent the consumption of the person taxed so investment is reduced less by a tax than by the deficit. That is the savings/investment story.
The sectoral composition of demand story is that a deficit leads the Fed to have somewhat higher interest rates than otherwise and those attract some foreign capital inflow whihc "strengthens" the dollar reducing the profitability of producing tradeable goods compared to non-tradable goods => more imports and fewer exports of manufactures and agriculture.
“But if the problem of how to set up a factory has been mostly solved, it might put a big damper on overall manufacturing productivity growth.”
I think an equally important problems have yet to be solved: location of manufacturing facilities and getting the most out of products/technologies. Too often, government incentives and corporate strategies are one-dimensional when they should be multi-dimensional. Two glaring examples are new chip fabs. The U.S. and its allies are well position for leading-edge chip production. The parking of thousands of newly manufactured vehicles during the pandemic is but one illustration of the importance of trailing-edge chips. Instead, the government is throwing money at INTL and TSMC to manufacture leading-edge chips in the worst-possible location: the desert of Arizona. Hundreds of millions of gallons of fresh water will be annually wasted, even if INTL’s claim of recycling 95% of water used in its chip fab is true. Hydropower is on the wane because of climate change-induced extended droughts and record heatwaves, to say nothing of increased use of air conditioning (both industrial and residential). The last thing Arizona needs is an increased waste of fresh water, increased use of electrical power, and more industrial and residential development (housing for jobs generated by chip fabs). The government should have used subsidies to incentivize the type of chip fabs as well as their geographic locations: trailing-edge chips fabs in the Rust Belt/Great Lakes Region. The largest source of fresh water, cooler temperatures to reduce cooling needs/costs for chip fabs, and jobs for a region that is long overdue for a job renaissance.
Solar and wind energy farms should also be incentivized by the government. It makes no sense to build-out nuclear and additional hydropower dams. Nuclear power development is outrageously expensive and takes a very long time to deploy when compared with alternatives. Costs for solar energy continue to rapidly decline, as well as the cost of leading-edge battery design/technologies. Proliferation of nuclear power would the world’s worst unforced error, saddling millennia of future generations with the challenge and danger of managing nuclear waste., to say nothing of providing targets for terrorists. Solar/wind and agriculture benefit each other. The government is subsidizing 30 million acres for growing corn for ethanol. Ten million acres for deployment of solar panels is enough to supply all of the U.S. electricity demand. Furthermore, solar panels are up to 20% more efficient when deployed above greenery. Solar panels would result in a significant savings of fresh-water irrigation lost to transpiration. Alternative crops could be grown at current or greater yields for the amount of fresh-water used. There is a reason Iowa is a leader in Alt-Green Energy. You present a farmer with more opportunity to generate more income per acre (wind and solar farm leases) and increase crop yields while dropping his/her cost for irrigation and there will be a lot of Iowa stories throughout the Grain Belt.
In essence, it’s important to be increasing manufacturing in a number of sectors, but it should be done in a way that has multiple positive, productive associated benefits via wise use of government incentives and resource conservation. In a world that will experience worse climate-change effects in the near future, we can’t afford to keep using a one-dimensional strategy.
These are not pie-in-the-sky incentives. Producing more from our limited natural resources is always important, but especially important in a world that will face increasing resource challenges because of climate change.
Growing lettuce and alfalfa in the blistering Western US sunshine is pure foolishness, especially when you then export said crunchy water from water-scarce places in order to feed cows. This video from Vox is an excellent dive into that particular policy facepalm: https://youtu.be/f0gN1x6sVTc?feature=shared
Relying on big, open lakes to store water and produce hydropower in the desert: also dumb. Less silly when you're at least putting solar panels on the lakes to stack functions, but still not a great system for water management in a doggone arid environment. Hubris...
Growing crops that are either A) well suited for dry conditions (like corn, a C4 crop that is efficient in its water is), or which are in an agrivoltaic setup like those you describe, is just good ag and water policy. Same goes for switching away from hydropower and any steam-driven power source (nuclear, coal, gas) unless that steam is also being used as a way to purify/distill water..
Same goes for chip fabs, and other water intensive industries. Put 'em where water isn't a key constrained resource.
At least solar power is consistently abundant in AZ, so the hydropower shortage should only be a short term speed bump in net energy availability.
So many opportunities to do things smarter in this country (and many others)... at least all this gives me plenty of hope re: the abundance of opportunities out there for smart folks to make positive impacts (and get paid to do it!) so long as policymakers and corporate execs don't prevent the work from happening because it isn't beneficial to incumbents.
To be really clear: 1/3 of the water used in the entire Western US is for animal feed. Mostly for cows. (Source: that Vox video linked above). So, much as I agree with not putting water intensive chip fabs in the desert... Not growing alfalfa and hay there to feed cows would remove a *huge* part of that constraint.
If current megadrought trends continue there won't be water for chips -or- cattle feed. I'm not sure about current percentages but in the glory days of globalism we were shipping a lot of cattle feed overseas in super cheap return trips of container ships which is complicated and wrong on soooo many levels. (Arizona is ending Saudi leases, not sure if that's uniform or selective).
The reality is, we've treated ground water as inexhaustible and that particular party is coming to an end, accelerated by climate change. Expect much more farm disruption in the Plains states (along with accompanying social impacts) in the future.
The entire Oglala Aquifer--basically every well between the Rockies and the Mississippi--is going to run dry, due to massively unsustainable extraction rates. Ditto for the entire Colorado River basin. In a three-way water war between population/manufacturing/agriculture, I predict livestock-feed agribusiness will bite the dust in the early rounds.
Like as not, the farms not directly supplying consumers will all go bust, and the prairie will return, populated by grass-fed cattle/buffalo ranches.
Yeah, that's basically what I'm citing. And storage equals recharge-extraction and climate change has torpedoed recharge.
It's facile to say that arlg is going to "lose" since the cheap food it underpins is the foundation for social stability, not to mention hundreds of thousands of rural jobs. And those people are already unhappy.
Livestock feed for cattle is actually very expensive "finishing" process; i.e., "corn-finished beef". Ironically, it also destroys much of the nutritional value of beef, degrading a 2:1 Omega-3: Omega-6 ratio, down to 1:4...or worse.
But Cargill and ADM are very politically powerful corporations, that along with Monsanto will push strong for Big Ag. But none of them can refill the aquifers...
So, the interesting sub-question - WHY do large corporations, etc. do the dumb things that you point out? Perhaps there are other advantages to a location, but I'd bet more on politics and subsidies from local / state governments.... (or externalities that are effectively subsidies - ex. cheap water in a very dry place).
It would be amusing, if not so unhappy -- the new solar+ag movement seems also to have attracted marginal players, grifters, and PE. Farmland has ludicrously bad ROI so it is ripe for schemes that promise extra income. The solar+ag, hvdc, etc. involve contracte that guarantee disruptive access to panels and transmission facilities (maintenance, right?) and some come under the threat of eminent domain.
The large scale industries employ armies of project managers who talk farmers into deals that may be very undesirable -- I've heard of one scheme where the contract gives the company the right to *borrow against the land they are employing*. The average age of a farmer in the US is 58 -- these owners are old people, often using tenants or lease agreements and many of these transactions look a lot like boiler rooms.
Land is seen as a hedge against future economic disruption bidding the price up, making it inaccessible to young farmers. When those old farmers retire or leave the scene, a new crisis is in store.
Good to know about, those are some nasty contractual clauses and bad faith actors to be aware of. And I'm glad you took the tangent; I'm working on a local implementation of EPA Greenhouse Gas Reduction Fund financing programs, and it's important to understand the shady side of this particular industry sub-sector.
It's knowledge about the poor/usurious side of things that enables us to do things better, and shift the norms in the right direction through financial incentives, outreach, and marketing/PR. In my experience there are a lot of above-board solar developers who have to work overtime to make up for the smaller percentage of (really) bad actors out there.
I live in Caterpillar country. We have lots of factories here because of it. Everyone I know who works within this ecosphere is so busy they’re about to drop.
What is definition of output in these scenarios? GDP by sector?
Tax scams (moving IP to Ireland, Lux et al) has reduced domestic product (and inflated EU GDP).
Also transfer pricing- generally companies wish to minimize domestic earnings in US and allocate as much product revenue as possible to foreign sources, reducing domestic product
Just throwing that out there. Very interesting piece (and charts).
I'm going to declare for Hypothesis #3. Why? Because productivity is basically a logarithmic learning curve. 1 10 100 1000. Going from the 100 millionth foot of glass or cars made to a billion made takes a while. All the big productivity gains are early 1 to 100k to 1M. There is pyramid of things from big scale, continually made to batch job to discrete. Volume still rules to achieve productivity continually. Game changing tech can break paradigms.
Some thoughts about manufacturing productivity from experience.
A. is different than rapid batch manufacturing like plastic molding, progressive die stamping; glass bottle blowing;
B. Is different than long continuous but discrete run manufacturing, like wire drawing and fiber extrusion;
C. Is different than continuous flow process manufacturing like float glass sheet production, polyester and polythene plastic production, or gasoline.
2. Not all manufacturing production uses, per se, machines. Albeit its all "mechanical and not software"
3. Manufacturing productivity is generally based on the established learning curve theory. Its a century old and half a century older concept than Moores law.
4. Such productivity is idle and static, without a determined applied force to seek and develop small incremental improvements. Capital productivity occurs more dramatically when a large leap is made to a new manufacturing process to make a current thing.
5. Things that are made by manufacturing fall into a hierarchy that rests on the foundations of:
a. Extraction of raw materials
b. Conversion to refined discrete usable materials
c. Conversion of these materials into suitable forms for well form making processes. Examples - Titanium into machinable rods, steel into sheet coils, plastics extruder into pellets (non Rabbit eating pellets!!),
d. Components are manufactured from the above: fenders, wiring, connectors, laptop upper and lower case, water bottles, fabrics, light bulbs, cell phone camera lens
e. Components merge and are brought together and organized to make sub systems - wiring, connectors, lighting and switches into a wire harness for a car or computer. Circuit boards have resistors, capacitors, heat sinks, IC, CPU, memory added
f. Sub-Systems are joined up to make systems, like an electric dashboard with lighting, driver controls, with displays.
g. Finally, Product assembly - Final automotive assembly plants where robots weld and paint a car a minute ideally; Samsung and Foxconn assemble your cellphone at a rate of maybe every 10 to 30 seconds.
So... who drives productivity in these various areas?
It is first the incumbents. But sometimes its the game changer. Example is Cathode Ray Tubes, the CRT (some may never have one). CRTs were great tech in their time. But RCA, Zenith, GE tv guys were kinda slow on the development of the fluorescent edges, TFT 3 filter based flat panels coming from Toshiba, Sony (who survived the transition). Plasma and TFT were 1960s tech. But took to about early 2000s 40 years later to begin broad penetration
I just happened to have a long career in engineering plastics. Used in so many types of products. And, also learning all about all the other materials and processes.
I asked a manufacturing exec this question. Their answer is we have exported all the high volume manufacturing out of the US. What is left (in broad strokes) is super niche, complex, and low volume. So there is low value in investing in additional tech for these types of niche low volume products.
It's less that you don't need them than the fact that mass customization and complexity generally are hard (expensive and require smarter people) to automate. and those smarter people... are mostly not going into a low pay field.
It's possible that America's supply chains aren't big enough. In Asia you have the largest and most spread out supply chain on the planet. America's manufacturers can only spread to Mexico - a country beset by their own disfunctions. The rest of the Latin American are even big basket cases.
One thing you did not mention was that the infrastructure of the US is ageing and failing. This will lead to delays and additional cost of transportation. How many sectors would greatly benefit from having a rail line both industrial to transport goods and passenger to bring in the workers.
You also have the fact that those at the top are vastly over paid. Even if the monies didn't go directly to people actually working it would allow for better investment if facilities, machinery and general upkeep. I don't know how many times some equipment fails but rather that purchase new, high quality pieces that will last a long time they either buy the cheapest product they can or just makedo. Same with things like AC that computer systems rely on, but it is deemed 'too much' even though they will make up for it rapidly. But, its okay to pay a CEO millions for sitting on his rectum.
Management plays a key part here, a business can not really afford to have a bunch of brown noses just bobbing their heads. If they kept those in line and to a minimum things would actually progress at a far more rapid rate, but, too much work I guess.
Again, I suspect that short term stock market "thinking" drives a lot of bad investment decisions. Heck, even in the software world, it drives a ton of bad "cheap" hires that are far more expensive over time (what I would call "personnel debt" as analogy to "tech debt"
General Motors at one point had fourteen--yes, 14--layers of management, creating massive waste and inertia. My first car buy in 1981 was going to either be a Chevy Chevette or a Honda Civic (the first model after the CVCC). Both cost the same; but the Chevy (which I'd become familiar with at work) was a clunky piece of shit by comparison.
But Chevy--and Detroit in general--kept blaming the ass-kicking they were getting from the Japanese on the unions. When all along, even after NAFTA started slashing their costs, their greatest problem was managerial bloat and incompetence.
As a non-economist 😉 I have thoughts that both the injection of private equity into so many sectors including manufacturing is crippling us because it is completely parasitic. The incentive for stock buyback also disincentivices investment in robust R&D and machine investment and far too many companies and gobbling up money in stock buy backs. Monopolization is also a huge disincentive to productivity growth as it also cripples R&D and other necessary growth investment. So long as the US allows these three things to run the way they are currently America's productivity will continue to decline until we become too top heavy and the entire system crashes down. But that just my own opinion.
How many trillions have been used by companies for stock buybacks instead of investing that money in upgrading manufacturing? That investment in boosting stock prices will vanish, or not rebound for a decade or two, when this bubble ridden market finally resets.
I think a lot of what we see is the expansion of the wealth pyramid -- if you have more people every year and constant or even slight increase in gdp per person, there's an overwhelming amount of wealth that can be directed towards... something...globally.
I think, as much as anything, the buybacks are a symptom of companies not having new ideas for value creation proportional to the cash they're reaping.
Add in the effects of monopoly and monopsony practices and you get a lot of our best innovative energy being spent towards financial engineering and lowering expenses as the choice of how to increase profits.
One of the things that I like about your blog is that you are critical of economic data. So many people go to extremes these days either worshiping the chart or rejecting all statistics. Maintaining a judiciously critical attitude is the difficult position.
Another thought that popped into my head while reading this post is the fact that military effectiveness demands both constant improvements in equipment and human capital. Many are saying that we have over invested in gigantic ticket items, such as fighter planes, aircraft carriers, and tanks--while under investing in low ticket items, such as drones. What we must absolutely be best at is coordinating material and human beings so that we can find out where to attack and concentrate our forces on it.
I think that often the back seat reformers underestimate the capabilities of those big ticket items. After all, sufficient investment in EW capabilities may make those drones completely innefectual in a few years. People are often wrong about which machines will be useful in the future. The Bradley was ridiculed mercilessly yet it seems to be one of the most useful vehicles we have sent to Ukraine.
But the Abrams tanks were withdrawn from the front lines because they were too vulnerable.
Pretty much all MBT's like the Abrams and Leopard II are considered more useful as long-range snipers, doing shoot-and-scoot. Bradleys are lighter, more agile, and more numerous; a combo of reasons they show up more often on the forward envelope of engagements.
Thank you for all that useful information, but it doesn't change the fact that the Ukrainians withdrew the Abrams from the frontline because of their vulnerability.
My info is clearly out of date.
We sent 31 Abrams, 5 of which have already been lost; while we've sent 252 Bradley's. Ukraine has also lost most of its 90 Leopard II MBT's in combat, as well as some of its Challengers. My *guess* is the surviving MBT's are being kept in rapid reserve units outside of drone range.
The survivability of MBTs is evolving so rapidly that it is impossible to know where we will be in 2025. But since they were designed to be used in warfare of maneuver, it's no surprise that they are not much use in positional warfare.
The most important aspect of military effectiveness is having the social cohesion for sustaining a war long enough to win.
Afghanistan proved this can be done (even if the goals were wrong).
On the other hand, Ukraine and Israel are showing increasing politization/polarization of "defense of democracy". It seems only wars started by the right will receive public support. The right's isolationism is a military problem.
Afghanistan was a prime example of an unwinnable war. It was an unsustainable $90bn/year GOP-created money pit, that collapsed practically to the Kabul city limits only a few months after the Trump Admin withdrew air support & logistics to the Afghan Army.
The notion that Afghanistan was "winnable" is entirely due to the bullshit statistics the DOD released in an endless brown torrent of slime through every presidential administration after 9/11, where "winning" existed only on paper.
I'd like to see a deeper dive into where this data comes from actually. It's good that noah points out the possibility of data problems but articles about GDP or productivity trends are ten a penny and always seem to gloss over where the data comes from.
When people spend more time writing about manufacturing productivity instead of doing manufacturing....
Love the blog, but I think you're overlooking Learning Curves (Wrights Law) as the primary factor behind declining manufacturing productivity. These curves state that production costs drop significantly with each doubling of units produced, but only for standardized items made in factories.
This explains why housing and solar costs remain high (not factory-made), Moore's Law exists (it's just Wrights Law for chips), and market share matters (more volume = lower costs).
Most U.S. manufacturing has shifted from high-volume factory production (where Learning Curves thrive) to construction, repair, and custom work—none of which benefit from these curves.
Case in point: Dayton, Ohio. In 2007, the city have over 400 machine shops, many producing standardized parts for giants like GE, NCR (National Cash Register), and Delphi Auto. However, Delphi left in 2005, GE in 2008, and NCR in 2009. Today, less than 100 shops remain, primarily focused on repair work, custom projects, and small-batch manufacturing—work that doesn't benefit from learning curves. If you were to analyze Dayton's manufacturing productivity today, it would likely show a steep decline, but this is because the nature of the work has changed. Volumes matter.
Learning curves are also the reason I didn't go bankrupt. In 2016, I bought a struggling equipment manufacturer. When a supplier threatened to bankrupt me, I tried making their product myself. My costs were 6x higher than theirs. However, a 15% learning curve predicted I'd break even at 80 units and match their costs at 2,000. I went all-in, and within 3 years, we were more profitable than ever. No amount of money or technology could have sped up that process; we simply had to produce repeatedly until we became efficient.
Learning Curves are real, but they only apply to specific types of production. The U.S. has largely moved away from those types, which is the real reason behind the decline.
If you want more info, there are good sources here:
https://www.construction-physics.com/p/where-are-my-damn-learning-curves
https://www.ark-invest.com/wrights-law/
https://hbr.org/1964/01/profit-from-the-learning-curve
https://www.bcg.com/publications/1968/business-unit-strategy-growth-experience-curve
That was a typo on solar.
Housing and *construction* costs remain high because it’s not factory made. Solar is taking over the world :)
Learning curves are also why solar is going to beat wind because more of its BOM is in the factory, wind has higher transportation and construction which doesn’t benefit the same.
Then the next question is why the US has moved away from production that benefits from learning curves.
As an entrepreneur who started two high tech manufacturing companies, I agree that the availability of capital is challenging for manufacturing capital goods. We did it, but we were the 'odd duck' in that we used capital for manufacturing assets. I agree that the capital markets in the US are not oriented towards such investments, but could be encouraged by government policy like tax incentives for manufacturing investments.
I wonder if you combine the limited capital available for manufacturing and the use of profits for stock buyback, then you see a big effect. Hard to get big manufacturing productivity gains without easy capital access. Look at how easy it has been for quite a while to get capital for various software ventures (the Silicon Valley VCs, also the current AI hype).
Now, it is possible that the lack of capital is because sources don't trust the profitability of manufacturing, but also quite likely that it doesn't show the ROI immediately enough to satisfy today's very short term stock market thinking (and IPOs)
Gary,
One way to look at it is that plenty of money is also invested in bonds, which does not provide a return above the interest rate (unless purchased distressed). Capital goods should be financed from loans or other debt-like structures, with the main concern here by investors is the security of the principle. Here is where government incentives or alternate structures could help. A key aspect is pooling loans across a portfolio. Then tranches can be separated, analogous to how this is done in mortgages. this may be a bad idea: maybe a Freddie or Fannie for manufacturing investments? Another way this could work is for government absorbing first losses from the portfolio, thereby reducing risk. In any case, the government has figured out how to do some version of this, as seen by the success of the TARP program and the energy loan programs. So, there is a pony in there somewhere.
The electrons vs. atoms duality also biases for electrons (tech) because atoms (manufacturing of material things) are asset-burdened enterprises.
Theoretically, a tech company can maneuver faster because it doesn't have to buy or sell property, plant or equipment. Everything a manufacturer does has to be done slowly, even for a declining unit. It can take several quarters to go from the decision to close a plant to actual shutdown (disposition of labor, finding a buyer or salvage value for labor, selling the property, possibly being required to remediate the property in order to be sellable, etc.). The investors feel, "If I can move -illions of dollars with the click of a mouse, so should the companies I invest in."
Even if we continue to tax business income -- which we should not -- at least we should allow expensing of investment. [This would also have the advantage of eliminating the distortion of faster depreciation for some sectors and assets and undermine investment tax credits for certain activities.]
Why we should not?
Why eliminate the distortions? They lead to investing more or less in the favored/disfavored activities rather than in those that maximize income. If youmean tht the incentive effects ae good, then it's probably because it is good to have more OUTPUT of the favored sector, not more INVESTMENT, so the incentive should be directed at the output.
From an economic point of view you may be right, but countries are thirsty for money. The trend is for all companies to pay at least 15% corporate tax, including those based in what used to be called tax havens.
?? And I want to quench some of that that thirst by taxing business income as personal income or better as personal consumption.
If all corporations were structured as LLCs, then that might fly. It just seems reasonable that, for example, a US business benefits tremendously from the strength of the US military and the resulting stability in the world, and somehow businesses should help pay for that privilege in due course, as every other person/entity does as well. If that were via a flow-through entity and paid through personal income taxes, I can see that (though I would really hate paying tax on non-cash income). In the end, as imperfect as the government is, we (Americans) derive tremendous benefits from its existence. It does need to be funded, and those in charge need to be held accountable. Very imperfect, but what a country we have the luck and privilege to live in.
Well, the OCDE (which is the leader in these changes) is very powerful, and this organization will not change its mind.
I'd put that as Hypothesis 5: Lack of the right kind of money that would invest in the fundamental breakthroughs required for automation.
Current automation, basically everywhere is based on one form of optical means or another, ranging from cameras to LiDAR.
Would policies/laws that discourage stock buy-backs help in this regard?
Stock buybacks are essentially a return of cash (capital) to stockholders, analogous to dividends but obviously quite different in effect. They are also a way of funding generous stock option incentives for executives (a way to juice compensation), and thus there is a personal interest by managers in making this choice in use of capital. That being said, returning cash to shareholders is an indication that a manager has no greater productive use of that cash. By increasing incentives for investment in the business, it would increase the amount of capital used for that purpose. However, discouraging buybacks would best be done through other means.
You mentioned overregulation of land use, and I think it’s highly probable that it hurts industry productivity specifically because it reduces “circulation of workers between manufacturing companies.”
Industry in the US has been spreading out, especially out of the northeast and Midwest and into the sunbelt, and its logical to expect that industries that scatter themselves among accommodating sunbelt towns rather than concentrating in industrial cities will have less circulation of workers.
Is this really true that American industry wasn't spread out before?
All I really know about is what I read in the book Janesville and I got the impression that it was pretty common for US manufacturing in early- and mid-century to be scattered among accommodating midwest towns rather than concentrating in industrial cities.
I mean, that's where the whole concept of "The Rustbelt" comes from, isn't it? That small cities like Flint and Janesville used to have high paying manufacturing jobs despite having populations under 100,000.
A quick Google finds this article[1] which claims "During the 20th century, America built thousands of manufacturing plants in small cities in the Midwest" in places like Danville, Newton, and Youngwood.
[1]: https://www.industryweek.com/talent/article/22028380/the-abandonment-of-small-cities-in-the-rust-belt
Yup. Lots of former mill towns all over rural America, I've lived in or worked in quite a few from the Northeast to the Southeast.
The movement to the sunbelt could also reduce labor productivity. Companies have been moving manufacturing jobs from high-wage union states in the North to low-wage non-union states in the South. It makes financial sense to move South even if your new workers in those states are less productive.
that may also lead companies to substitute capital for labor. It may be worth investing in industrial robots to make high-wage workers more productive but not worth the investment in lower wage areas.
But doesn't (modern) manufacturing need increasingly educated/trained workers? The sorts of employees produced from high-end Polytechs found in larger cities? I mean virtually all machining, for example, now seems to be computerized. The days where it was all done by hand by goes with only a HS diploma is long gone.
High wages in US tech/finance sector is really crowding out manufacturing
Is there any possible way to construct a two-speed America, like some envisioned for Europe about 20 years ago? Maybe some states can voluntarily give up US dollar so that they can do manufacturing on the cheap?
I don't necessarily disagree, but at the same time doesn't that suggest a surprisingly small pool of talent? Does that mean that even a country as large as the US can only be outstanding in a handful of areas? To be fair, I think it goes beyond tech and finance to also encompass life sciences, maybe healthcare, and maybe media, although I'm not sure how much media jobs would compete with manufacturing ones for talent.
Another hypothesis is there has been falling demand for US manufacturing. This has been driven in large part by the increased competitiveness of manufacturing first in Japan, then Korea, Taiwan and of course China.
Less demand for US manufactured goods made investments into new machine tools and equipment less compelling. With the benefit of hindsight, I'm not sure how the US could have kept many manufacturing industries alive here. However, we almost certainly could have done more to incentive more advanced manufacturing to stay.
For example, what would it have taken for Apple to produce iterations of the iPod and iPhone here? The iPhone is an extremely underrated reason for China's supply chain strength in consumer electronics that has spilled over in everything from EVs to drones and now eVTOLs.
I mentioned this in the "investment" section! :-)
From my read you looked at the customer side of the investment - we also have issues getting funding on the supply side. Automation will require a lot more breakthroughs to become economical.
An article on that subject in the NYT, years ago, claimed that Chinese industry can hire hundreds of engineers overnight. In the USA companies can’t hire _one_ engineer overnight. The background check takes two weeks.
Might be valid, but hard to see how that can have a major long term effect - the engineer still gets hired. Just makes China more nimble. To have a long term effect, we would need to assume that the slight nimbleness creates a big advantage. Not defending the slow hiring, just don't think it makes a big impact.
Every little bit hurts.
Add the effect of payroll taxes, and especially payroll tax accounting. The paperwork for a dishwasher who works for a week is almost as expensive as for a skilled technician who works the whole year.
I thought you could outsource payroll accounting, and save bigly.
Lower fiscal deficits would have helped.
Why is that? My understanding is that debt encourages foreign countries to buy our exports, although I don't really understand the casual mechanism.
You must be thinking of Ex-Im Bank credits. Sure that kind of credit can promote exports. But the macroeconomic machoism is that a deficit means, mainly, someone bought a USG bond instead of a private sector bond and the private sector bond was more likely to finance investment than consumption. So deficit transfers resources from investment to the expenditure. A tax reduces to some greater extent the consumption of the person taxed so investment is reduced less by a tax than by the deficit. That is the savings/investment story.
The sectoral composition of demand story is that a deficit leads the Fed to have somewhat higher interest rates than otherwise and those attract some foreign capital inflow whihc "strengthens" the dollar reducing the profitability of producing tradeable goods compared to non-tradable goods => more imports and fewer exports of manufactures and agriculture.
“But if the problem of how to set up a factory has been mostly solved, it might put a big damper on overall manufacturing productivity growth.”
I think an equally important problems have yet to be solved: location of manufacturing facilities and getting the most out of products/technologies. Too often, government incentives and corporate strategies are one-dimensional when they should be multi-dimensional. Two glaring examples are new chip fabs. The U.S. and its allies are well position for leading-edge chip production. The parking of thousands of newly manufactured vehicles during the pandemic is but one illustration of the importance of trailing-edge chips. Instead, the government is throwing money at INTL and TSMC to manufacture leading-edge chips in the worst-possible location: the desert of Arizona. Hundreds of millions of gallons of fresh water will be annually wasted, even if INTL’s claim of recycling 95% of water used in its chip fab is true. Hydropower is on the wane because of climate change-induced extended droughts and record heatwaves, to say nothing of increased use of air conditioning (both industrial and residential). The last thing Arizona needs is an increased waste of fresh water, increased use of electrical power, and more industrial and residential development (housing for jobs generated by chip fabs). The government should have used subsidies to incentivize the type of chip fabs as well as their geographic locations: trailing-edge chips fabs in the Rust Belt/Great Lakes Region. The largest source of fresh water, cooler temperatures to reduce cooling needs/costs for chip fabs, and jobs for a region that is long overdue for a job renaissance.
Solar and wind energy farms should also be incentivized by the government. It makes no sense to build-out nuclear and additional hydropower dams. Nuclear power development is outrageously expensive and takes a very long time to deploy when compared with alternatives. Costs for solar energy continue to rapidly decline, as well as the cost of leading-edge battery design/technologies. Proliferation of nuclear power would the world’s worst unforced error, saddling millennia of future generations with the challenge and danger of managing nuclear waste., to say nothing of providing targets for terrorists. Solar/wind and agriculture benefit each other. The government is subsidizing 30 million acres for growing corn for ethanol. Ten million acres for deployment of solar panels is enough to supply all of the U.S. electricity demand. Furthermore, solar panels are up to 20% more efficient when deployed above greenery. Solar panels would result in a significant savings of fresh-water irrigation lost to transpiration. Alternative crops could be grown at current or greater yields for the amount of fresh-water used. There is a reason Iowa is a leader in Alt-Green Energy. You present a farmer with more opportunity to generate more income per acre (wind and solar farm leases) and increase crop yields while dropping his/her cost for irrigation and there will be a lot of Iowa stories throughout the Grain Belt.
In essence, it’s important to be increasing manufacturing in a number of sectors, but it should be done in a way that has multiple positive, productive associated benefits via wise use of government incentives and resource conservation. In a world that will experience worse climate-change effects in the near future, we can’t afford to keep using a one-dimensional strategy.
These are not pie-in-the-sky incentives. Producing more from our limited natural resources is always important, but especially important in a world that will face increasing resource challenges because of climate change.
This. On so many levels.
Growing lettuce and alfalfa in the blistering Western US sunshine is pure foolishness, especially when you then export said crunchy water from water-scarce places in order to feed cows. This video from Vox is an excellent dive into that particular policy facepalm: https://youtu.be/f0gN1x6sVTc?feature=shared
Relying on big, open lakes to store water and produce hydropower in the desert: also dumb. Less silly when you're at least putting solar panels on the lakes to stack functions, but still not a great system for water management in a doggone arid environment. Hubris...
Growing crops that are either A) well suited for dry conditions (like corn, a C4 crop that is efficient in its water is), or which are in an agrivoltaic setup like those you describe, is just good ag and water policy. Same goes for switching away from hydropower and any steam-driven power source (nuclear, coal, gas) unless that steam is also being used as a way to purify/distill water..
Same goes for chip fabs, and other water intensive industries. Put 'em where water isn't a key constrained resource.
At least solar power is consistently abundant in AZ, so the hydropower shortage should only be a short term speed bump in net energy availability.
So many opportunities to do things smarter in this country (and many others)... at least all this gives me plenty of hope re: the abundance of opportunities out there for smart folks to make positive impacts (and get paid to do it!) so long as policymakers and corporate execs don't prevent the work from happening because it isn't beneficial to incumbents.
To be really clear: 1/3 of the water used in the entire Western US is for animal feed. Mostly for cows. (Source: that Vox video linked above). So, much as I agree with not putting water intensive chip fabs in the desert... Not growing alfalfa and hay there to feed cows would remove a *huge* part of that constraint.
If current megadrought trends continue there won't be water for chips -or- cattle feed. I'm not sure about current percentages but in the glory days of globalism we were shipping a lot of cattle feed overseas in super cheap return trips of container ships which is complicated and wrong on soooo many levels. (Arizona is ending Saudi leases, not sure if that's uniform or selective).
The reality is, we've treated ground water as inexhaustible and that particular party is coming to an end, accelerated by climate change. Expect much more farm disruption in the Plains states (along with accompanying social impacts) in the future.
The entire Oglala Aquifer--basically every well between the Rockies and the Mississippi--is going to run dry, due to massively unsustainable extraction rates. Ditto for the entire Colorado River basin. In a three-way water war between population/manufacturing/agriculture, I predict livestock-feed agribusiness will bite the dust in the early rounds.
Like as not, the farms not directly supplying consumers will all go bust, and the prairie will return, populated by grass-fed cattle/buffalo ranches.
Yeah, that's basically what I'm citing. And storage equals recharge-extraction and climate change has torpedoed recharge.
It's facile to say that arlg is going to "lose" since the cheap food it underpins is the foundation for social stability, not to mention hundreds of thousands of rural jobs. And those people are already unhappy.
Livestock feed for cattle is actually very expensive "finishing" process; i.e., "corn-finished beef". Ironically, it also destroys much of the nutritional value of beef, degrading a 2:1 Omega-3: Omega-6 ratio, down to 1:4...or worse.
But Cargill and ADM are very politically powerful corporations, that along with Monsanto will push strong for Big Ag. But none of them can refill the aquifers...
https://phys.org/news/2023-06-weve-groundwater-nudged-earth.html
So, the interesting sub-question - WHY do large corporations, etc. do the dumb things that you point out? Perhaps there are other advantages to a location, but I'd bet more on politics and subsidies from local / state governments.... (or externalities that are effectively subsidies - ex. cheap water in a very dry place).
It would be amusing, if not so unhappy -- the new solar+ag movement seems also to have attracted marginal players, grifters, and PE. Farmland has ludicrously bad ROI so it is ripe for schemes that promise extra income. The solar+ag, hvdc, etc. involve contracte that guarantee disruptive access to panels and transmission facilities (maintenance, right?) and some come under the threat of eminent domain.
The large scale industries employ armies of project managers who talk farmers into deals that may be very undesirable -- I've heard of one scheme where the contract gives the company the right to *borrow against the land they are employing*. The average age of a farmer in the US is 58 -- these owners are old people, often using tenants or lease agreements and many of these transactions look a lot like boiler rooms.
Land is seen as a hedge against future economic disruption bidding the price up, making it inaccessible to young farmers. When those old farmers retire or leave the scene, a new crisis is in store.
(Sorry, this is all moderately off topic.)
Good to know about, those are some nasty contractual clauses and bad faith actors to be aware of. And I'm glad you took the tangent; I'm working on a local implementation of EPA Greenhouse Gas Reduction Fund financing programs, and it's important to understand the shady side of this particular industry sub-sector.
It's knowledge about the poor/usurious side of things that enables us to do things better, and shift the norms in the right direction through financial incentives, outreach, and marketing/PR. In my experience there are a lot of above-board solar developers who have to work overtime to make up for the smaller percentage of (really) bad actors out there.
I live in Caterpillar country. We have lots of factories here because of it. Everyone I know who works within this ecosphere is so busy they’re about to drop.
Because there's not enough people - or automation.
What is definition of output in these scenarios? GDP by sector?
Tax scams (moving IP to Ireland, Lux et al) has reduced domestic product (and inflated EU GDP).
Also transfer pricing- generally companies wish to minimize domestic earnings in US and allocate as much product revenue as possible to foreign sources, reducing domestic product
Just throwing that out there. Very interesting piece (and charts).
Yeah, it's value-added by sector.
I'm going to declare for Hypothesis #3. Why? Because productivity is basically a logarithmic learning curve. 1 10 100 1000. Going from the 100 millionth foot of glass or cars made to a billion made takes a while. All the big productivity gains are early 1 to 100k to 1M. There is pyramid of things from big scale, continually made to batch job to discrete. Volume still rules to achieve productivity continually. Game changing tech can break paradigms.
Some thoughts about manufacturing productivity from experience.
1. Batch/job setup manufacturing (like machining);
A. is different than rapid batch manufacturing like plastic molding, progressive die stamping; glass bottle blowing;
B. Is different than long continuous but discrete run manufacturing, like wire drawing and fiber extrusion;
C. Is different than continuous flow process manufacturing like float glass sheet production, polyester and polythene plastic production, or gasoline.
2. Not all manufacturing production uses, per se, machines. Albeit its all "mechanical and not software"
3. Manufacturing productivity is generally based on the established learning curve theory. Its a century old and half a century older concept than Moores law.
4. Such productivity is idle and static, without a determined applied force to seek and develop small incremental improvements. Capital productivity occurs more dramatically when a large leap is made to a new manufacturing process to make a current thing.
5. Things that are made by manufacturing fall into a hierarchy that rests on the foundations of:
a. Extraction of raw materials
b. Conversion to refined discrete usable materials
c. Conversion of these materials into suitable forms for well form making processes. Examples - Titanium into machinable rods, steel into sheet coils, plastics extruder into pellets (non Rabbit eating pellets!!),
d. Components are manufactured from the above: fenders, wiring, connectors, laptop upper and lower case, water bottles, fabrics, light bulbs, cell phone camera lens
e. Components merge and are brought together and organized to make sub systems - wiring, connectors, lighting and switches into a wire harness for a car or computer. Circuit boards have resistors, capacitors, heat sinks, IC, CPU, memory added
f. Sub-Systems are joined up to make systems, like an electric dashboard with lighting, driver controls, with displays.
g. Finally, Product assembly - Final automotive assembly plants where robots weld and paint a car a minute ideally; Samsung and Foxconn assemble your cellphone at a rate of maybe every 10 to 30 seconds.
So... who drives productivity in these various areas?
It is first the incumbents. But sometimes its the game changer. Example is Cathode Ray Tubes, the CRT (some may never have one). CRTs were great tech in their time. But RCA, Zenith, GE tv guys were kinda slow on the development of the fluorescent edges, TFT 3 filter based flat panels coming from Toshiba, Sony (who survived the transition). Plasma and TFT were 1960s tech. But took to about early 2000s 40 years later to begin broad penetration
https://spectrum.ieee.org/the-consumer-electronics-hall-of-fame-fujitsu-plasma-
This is *such* a useful and well reasoned, wonderfully wonky response. Thanks for putting the hard work in to make the comment area awesome!
Thanks. Love wonky !!
I just happened to have a long career in engineering plastics. Used in so many types of products. And, also learning all about all the other materials and processes.
I asked a manufacturing exec this question. Their answer is we have exported all the high volume manufacturing out of the US. What is left (in broad strokes) is super niche, complex, and low volume. So there is low value in investing in additional tech for these types of niche low volume products.
Why would you not still need machines for "niche" products?
Niche products sell in smaller volumes, so the machines you can afford are only less expensive ones.
Lower volumes, but higher value? I do not see a direct, unidirectional link between product composition and investment intensity.
It's less that you don't need them than the fact that mass customization and complexity generally are hard (expensive and require smarter people) to automate. and those smarter people... are mostly not going into a low pay field.
Not enough scale?
It's possible that America's supply chains aren't big enough. In Asia you have the largest and most spread out supply chain on the planet. America's manufacturers can only spread to Mexico - a country beset by their own disfunctions. The rest of the Latin American are even big basket cases.
One thing you did not mention was that the infrastructure of the US is ageing and failing. This will lead to delays and additional cost of transportation. How many sectors would greatly benefit from having a rail line both industrial to transport goods and passenger to bring in the workers.
You also have the fact that those at the top are vastly over paid. Even if the monies didn't go directly to people actually working it would allow for better investment if facilities, machinery and general upkeep. I don't know how many times some equipment fails but rather that purchase new, high quality pieces that will last a long time they either buy the cheapest product they can or just makedo. Same with things like AC that computer systems rely on, but it is deemed 'too much' even though they will make up for it rapidly. But, its okay to pay a CEO millions for sitting on his rectum.
Management plays a key part here, a business can not really afford to have a bunch of brown noses just bobbing their heads. If they kept those in line and to a minimum things would actually progress at a far more rapid rate, but, too much work I guess.
Again, I suspect that short term stock market "thinking" drives a lot of bad investment decisions. Heck, even in the software world, it drives a ton of bad "cheap" hires that are far more expensive over time (what I would call "personnel debt" as analogy to "tech debt"
General Motors at one point had fourteen--yes, 14--layers of management, creating massive waste and inertia. My first car buy in 1981 was going to either be a Chevy Chevette or a Honda Civic (the first model after the CVCC). Both cost the same; but the Chevy (which I'd become familiar with at work) was a clunky piece of shit by comparison.
But Chevy--and Detroit in general--kept blaming the ass-kicking they were getting from the Japanese on the unions. When all along, even after NAFTA started slashing their costs, their greatest problem was managerial bloat and incompetence.
An excellent exploration.
Question: what role might education play in this productivity puzzle?
As a non-economist 😉 I have thoughts that both the injection of private equity into so many sectors including manufacturing is crippling us because it is completely parasitic. The incentive for stock buyback also disincentivices investment in robust R&D and machine investment and far too many companies and gobbling up money in stock buy backs. Monopolization is also a huge disincentive to productivity growth as it also cripples R&D and other necessary growth investment. So long as the US allows these three things to run the way they are currently America's productivity will continue to decline until we become too top heavy and the entire system crashes down. But that just my own opinion.
How many trillions have been used by companies for stock buybacks instead of investing that money in upgrading manufacturing? That investment in boosting stock prices will vanish, or not rebound for a decade or two, when this bubble ridden market finally resets.
I think a lot of what we see is the expansion of the wealth pyramid -- if you have more people every year and constant or even slight increase in gdp per person, there's an overwhelming amount of wealth that can be directed towards... something...globally.
I think, as much as anything, the buybacks are a symptom of companies not having new ideas for value creation proportional to the cash they're reaping.
Add in the effects of monopoly and monopsony practices and you get a lot of our best innovative energy being spent towards financial engineering and lowering expenses as the choice of how to increase profits.
https://ourworldindata.org/grapher/global-gdp-over-the-long-run?time=1700..latest