Seen vs unseen. If one state raises its manufacturing "total labor burden" (minimum wages, liability, taxes, workman's comp, unemployment, regulations, etc.) it effectively outlaws low-productivity and the statistics will show an increase in local manufacturing productivity. There are plenty of scenarios in which, over time, total local output still increases as the population and overall growth increase, but total local consumption of that sector's goods increases far more, helped by lower-productivity imports, especially if those are paid for by selling debt and assets to the makers of those imports, and which can also pay for putting the formerly-above-zero-marginal-product labor force on various kinds of subsidies and welfare since the law makes them below-ZMP now. Hooray!
It's a bit of a strange argument. She says that if a domestic manufacturer produces by importing cheap intermediate imports, that should count as a trade change, not a productivity increase. But it is both.
Hm. To take Houseman's example to an extreme for illustrative purposes, suppose an American car manufacturer firm moves all its production except for final assembly (screwing on wheels and bumpers) to Mexico, and gets rid of most of its local workforce. (Such 'screwdriver factories' are not unknown, for instance in ex-USSR.) The set of intermediate inputs - the "car kit" - for the final product has become much cheaper, but I don't see how it is reasonable to treat this as a productivity increase by the manufacturer firm in any meaningful sense. It's the same factories, same equipment and work processes, just lower labor costs - lower for reasons totally unrelated to the manufacturer's business. To take the example to the opposite extreme, even if the manufacturer switches the supplier of a basic input such as steel to a cheaper one (foreign or not), why should that count as an increase in productivity? I can see that the manufacturer has discovered valuable information, and that might count towards a productivity increase, but it doesn't feel right. Isn't switching suppliers a red herring? If the manufacturer's existing supplier implemented a new process - increasing _its_ productivity - and reduced prices, the effect on the manufacturer would be the same (lower input costs) as if it had found a cheaper supplier. It is strange to consider the former as a manufacturer's productivity increase, and that makes the latter suspect also. It sort of double-counts the supplier's productivity increase.
Perhaps this is all just terminological confusion on my part, though, using layman's meaning of "productivity" rather than as a term of art in economics. But if the term of art is so far out of whack with the lay meaning of the word, that's hardly good for discussion in this area.
The dramatic increase in the world economy's technological complexity made the advantages of "specialization and trade" even more valuable than in history. This specialization has expanded to the point where a small business can have an international monopoly on a product that every country or customer can make themselves. Free trade allows all possible customers worldwide to take advantage of the lower production costs.
I had a market for "staged frozen zebrafish embryos" with ± 10 minutes of fertilization time, which allowed biological researchers to determine what genes are turned on and off during critical development stages. For example, on neural tube closure, we know which genes are required at what times to prevent spinal bifida from occurring. Remember, we are a vertebrate like zebrafish. I had zebrafish production tanks where I could get tens of thousands of eggs in a few minutes, whereas the research labs with only a few fish would have to spawn them repeatedly to obtain the samples. At a time when the minimum sample size required a thousand or so embryos, I had a market from the Max Planck Institute in Germany to the University of Singapore and every major research university in between. Total sales were not that large, but the amount of Ph.D. level research time saved was huge. My infrastructure cost was covered by my 10+ million zebra fish/yr sales, and Google ads allowed highly targeted research customers to find me.
Science and technology have grown in my lifetime from the time when a "Renaissance Man" who knew at least the boundary of all human knowledge about the world could exist—and I met some of them—to now, knowing the boundary of human knowledge in even a few narrow areas is a stretch. This changes the world, but also allows fools with good rhetoric to make foolish statements in all areas of real knowledge, from economics to biology, and even engineering. You can even get the Surgeon General saying N95 masks don't work when working-class people, where OSHA has been mandating N95 masks for half a century, knew they did.
A lot of the concern is with manufacturing employment: the good jobs problem. There is also a resilience issue: e.g. shipbuilding. The hegemon of the maritime order should have more ship-building capacity. Yes, the Jones Act is an issue, but efficiency uber alles is not the basis of a resilient polity.
Shipbuilding and steel production are two of the most protected industries in the United States, and are among the industries in the worst shape. Protecting people and businesses from competition rarely makes them perform better.
I agree, but most of the decline happened before WWI, when none of these factors were significant. Poor organization and management seem to be the primary causes.
according the article, foreign labor costs and foreign steel costs actually HAVE been big factors dating back to at least the 1860's.
The third factor is when a small island/maritime nation, which already has low labor costs and low steel costs, decided to absolutely devote itself as a matter of honor to having the best shipbuilding organization and management in the world, too.
When you have all three, you become the premier shipbuilder in the world.
The public discourse will always be about employment, but as economists we should be steel-manning the point about output, not misleading people the discussion.
I think the efficiency vs. resiliency paradigm is a good way to look at it. Does it make economic sense for me and my family to eat out every day? Maybe from an efficiency perspective (though I doubt it). Does it make sense that so many people ALWAYS eat out and literally can't cook? Nope.
There are two primary problems with the resiliency argument that I see:
1: Choosing WHAT we should be resilient to. What do we choose to specialize away from given that we can't know how to do everything. I make my own knives and garden tools, but obviously I am weird because I find it fun. Other people can't replace the fill and flush mechanisms on their toilets; I don't find that fun, but it is worth doing myself. The government stepping in to decide what jobs and industries are necessary for people to do would be a disaster, and if the gov. focused only on defense focused industries (which does seem reasonable) the impact on manufacturing would be miniscule. You just wouldn't see millions of people suddenly employed making ships, I think.
2: Arguing that the government will pick the right balance between efficiency and resiliency seems to ignore everything we know from experience about how government functions. Even assuming the government could do that in some fairly effective way, as opposed to finding a spot way inside the production possibility curve like they normally do, empirically the government does not seem at all interested in resiliency. Consider the many departments whose notional role is to maintain resiliency: the Federal Reserve, FEMA, everyone who is supposed to keep wildfires from gutting central CA, HHS, Department of Transportation... Generally the behavior of government is to shift as many resources away from resiliency and towards fancy new things as possible, because the incentives favor putting your name on fancy new stuff and not paying attention to the boring day to day operations. I would absolutely expect bureaucrats to optimize for favoring their buddies and flashy numbers at the expense of resiliency. When the crisis comes and their resiliency fails, best case for them they will be retired, and even if not, well who could have seen this coming? It is unprecedented!
Just in terms of our current problems, we didn't have to be over a barrel for rare earth magnets. The government absolutely could have made cut outs in the EPA regs for these important defense related items but, nah, why do that? The government protected the mohair industry from outside competition, but not the rare earth mining and refining industry from other branches of their own government. There is a lesson there.
Yeah, to be clear, I'm not arguing for government to come in and pick industries to subsidize.
Rather, I'm pointing out that, I think in large part, our current state of affairs is implicitly based on a lot of government subsidization. What's "efficient" for the market to produce is based on what the government subsidizes (or not) by regulatory policy. With respect to trade, it's "efficient" to import things because they're produced in ways that would be illegal to do domestically.
I still find that kind of "efficiency" reasoning morally repugnant. If slavery is legal in Y, and this makes trading with Y efficient, the answer isn't
a) legalize slavery in the US
b) subsidize US industry
c) make war on Y
It's simply
d) don't trade with Y
Now, in many cases that aren't so morally repugnant as slavery, the answer would be A, reduce the legal and regulatory burden so that what is illegal in the US but legal in Y becomes legal in the US.
More generally, we presume economic efficiency based on voluntary trade. But the greater the divergence in legal rights between trading partners, the less that presumption can hold.
I don’t think you are entirely wrong, but I also think most of our regulation goes well past the horizon of “morally acceptable” and into the territory of “crazy people made demands and idiot politicians followed up for votes.” We should be rolling back lots of regulations and getting government thumbs off the scales as much as possible, not adding a few more thumbs with tariffs.
Just to the point on slavery, slavery is a lot less efficient than people tend to think. Unless there is a huge dearth of workers compared to capital, a state that does not obtain in most of the countries people are worried about, there is not an advantage for slavery.
Just because resilience may be hard does not mean it does not matter. Mainstream economics is getting migration catastrophically wrong, particularly in Europe, because it pays no attention to resilience.
I am not saying resilience isn't important or does not matter. I am saying that I will believe government might be capable of choosing the right trade off between efficiency and resilience when they stop talking about resilience while actively doing things that make society less resilient. Getting government, which has both extremely limited ability to understand what is needed for resilience and almost zero incentive to find out, to serve this purpose is rather a fool's errand I am afraid.
As to economics on immigration, economists are getting it wrong because they are assuming away all those pesky details like "institutions and culture matter" and "whoops, we have a welfare state." Not to mention "we hate to actually enforce property laws, especially against brown people." I don't disagree that economists get themselves blinded by ideology now and again; the AEA is a bloody mess. That doesn't mean that one can say "resiliency matters, and I believe this will increase resiliency" and have it be true; it is worth discussing how that would actually happen with reference to institutions as they actually exist.
People will do almost anything for enough money, but free employment stops offering more money when it exceeds the value gained from the work; slavery does too, but the value paid includes guards, overseers, patrols, and other overhead, because people value their freedom. I don't know what economists call this, but to me, it means that slavery can never extract as much work from slaves as it could from free men. The only work which benefits from slavery is that which no free man would do, mostly because it kills its workers, reducing further how much training is worthwhile with slaves.
I don't see slavery as anything but control freakery.
Even when slaves were really well paid (eg salt miners in the UK) and only couldn't leave, they still had to be kept under lock and key (figuratively speaking in that case).
Your "Choosing WHAT we should be resilient to" is the main problem, to me. It assumes that "we" should choose such things, "we" should steer the economy instead of letting it take its own course.
That "we" can only mean government, and not only does government in general have a poor track record in everything they mismanage, but elections change priorities and policies so often that no choices last long enough to do anything but muddy the water.
It's like a bunch of drunks taking turns trying to steer a raft, as if they had any control over where the river goes. They are more likely to run aground or capsize it.
Companies want to survive, so they have an incentive to consider resiliency. What are the sources of their inputs? What are the threats to those sources? How can we best handle those risks?
But not the resilience of the system as a whole. Biological systems select for it. Eventually, they fail (all species disappear eventually) but lineages continue because resilience is what is selected for. Polities need to concern themselves with resilience.
Mainstream Economics is getting migration catastrophically wrong, particularly in Europe, because it pays no attention to resilience.
I don't understand what you're saying. I get the impression you think politics, ie governments, is capable of getting resilience right. Or, IOW, you think governments are smart enough and have enough data to manage economies better.
I don't understand how anyone can believe governments know more than people with skin in the game.
"The political problem of disaster preparedness is especially acute for the most useful form, disaster avoidance. The problem with avoiding a disaster is that success often renders itself invisible. The captain of the Titanic is blamed for hitting the iceberg, but how much credit would he have received for avoiding it? ...
"The prophet’s paradox can undermine public support for proactive measures. The very effectiveness of these interventions creates a perception that they were unnecessary, as the dire outcomes they prevented are never realized. Consequently, policymakers face a challenging dilemma: the better they manage a potential crisis, the more likely it is that the public will perceive their actions as overreactions. Success can paradoxically erode trust and make it more difficult to implement necessary measures in future emergencies. Hence, politicians are paid to deal with emergencies not to avoid them"
Indeed. The optimal state for politicians is for the disaster to happen in such a way that it isn’t obviously their fault, be minor enough that it is easy to fix yet be hyped into the worst thing since the Cambrian extinction, then to be seen solving the problem while mostly just talking about it and spending the political capital elsewhere.
I know this is meant metaphorically, but I am too literal-minded not to connect it to the news a few years ago that Del Monte was shuttering many of its canning facilities, including one that was important to farmers south of here.
The company said it was going “asset-light”. Since it’s a canning company I’m not sure what will be left … little plastic containers of fruit, I guess, as people eat more like toddlers now perhaps.
We are supposedly all such foodies now!
But the consumer who doesn’t know how to make use of canned goods, thriftily, and tastily as part of home cooking - is a stupid consumer.
And stupid consumers will never be good producers, or much else.
Thank heavens the government seems to have planned for that, with economists in its (academic) pay floating UBI.
Looking at economic aggregates can miss important details. When we no longer have adequate manufacturing capabilities in munitions, pharmaceuticals, semiconductor chips, as well as shipbuilding, we are vulnerable. And don't we owe something besides welfare to our working and middle class fellow citizens whose jobs are affected by imports of finished goods and cheap labor?
Much of America’s munitions are still made in outdated, Army-owned plants. Thankfully, these are finally being replaced with modern facilities. Military procurement has long been a national scandal.
The next major war is likely to center around cyberattacks, cheap drones, and AI-guided missiles — not just tanks and aircraft carriers. Unfortunately, the U.S. military continues to invest heavily in expensive, traditional weapons systems like carriers and manned jets, even though they’re increasingly vulnerable to cheaper, decentralized threats.
Many service-sector jobs — like logistics, software engineering, communications, and cybersecurity — are now essential to modern warfare. War today isn’t just about building tanks and planes; it’s about information, coordination, and adaptability — all areas where service-sector skills are critical.
Our real vulnerability isn’t a shortage of factory workers. It’s the government’s failure to adapt its spending and strategy to how fast warfare is changing.
The U.S. has 149 pharmaceutical manufacturing plants employing about 52,000 people. These plants produce more than half of the medicines consumed in the country. The leading foreign suppliers of medicines, by dollar value, are Ireland, Germany, Switzerland, India, and the Netherlands.
Less than 8% of America’s 166 million workers are employed in manufacturing today. Studies show that every manufacturing job “saved” by tariffs or trade barriers ends up costing about six or seven jobs elsewhere in the economy. Don’t we owe our working- and middle-class fellow citizens something better than protectionism, high prices, high unemployment, and welfare?
Most of your generic drugs, that amoxicillin you used in your last infection, come from China/India and a lot of the raw ingredients India uses come from China.
In fact faced with the prospect of sanctions the cost of generic drugs could be going up quite a bit. We can pay it, but of course in a war if wouldn't be a 100% markup but an infinity markup.
As I said, there are tradeoffs to be made. I don't think either an absolute protectionist or open position is great. It is not a matter of finding the one right solution with no downside, which doesn't exist.
Don't we owe something besides higher taxes and more expensive goods to our working and middle class fellow citizens whose purchases are affected by bans on cheap goods and better jobs?
I am by no means an enthusiast for egalitarianism, but I think getting us cheap goods needs to be balanced against the more serious social problem of un- or under-employed fellow citizens.
Balanced? By whom? Oh yes, the government, by busybodies who know more than the individuals concerned, and who change policies every few years for no reason other than to distance themselves from the predecessors.
Whatever factors do, they won't be the right factors, they will be political factors which are irrelevant or wrong for almost everybody but a few favored cronies.
Us arguing about what government should do is meaningless in any practical sense. The only sense left is the idealist sense.
I am a nobody. Let me reference a somebody . In a Fortune 2004 article Warren Buffett advocated for tariffs /a coupon system to deal with the trade deficit. He likened the situation as two islands where one island gradually is in debt and then ultimately gets owned by the other island. An interesting read and expands the discussion. Treasuries are one thing and hard assets are another.
I beleive economists have become too comfortable with their narratives on myth 1 and are glossing over important things they would highlight in other cases.
Output, rather than the number of workers is the appropriate measure of myth 1. In real dollars, manufacturing output has stagnated in the US.
You point out that manufacturing output has doubled since 1980. But that tells us nothing about the relative productivity of mfg. The economy has grown 8-10x in that period depending on whether you are counting for inflation.
To draw a simple analogy, although farming employment has fallen, we are more productive and better off because we can grow everything we grew in 1980 and more. A simple way to say it is agricultural GDP per capita has risen.
Manufacturing GDP per capita has fallen though. We literally cant make what we used to make.
I'd like to see some sources for those claims. Everyone else I've seen writing about this shows the opposite, that both manufacturing output and productivity have increased and are still increasing, although the COVID lockdowns scrambled the data for a while.
To me, it's telling that this data isn't readily available. It's kind of like with a lot of crime and immigration statistics. I don't think a lot of research and looking is done, because the results won't win you anything.
At first I just asked Google AI, but when I looked closely, it wasn't 100% clear the AI knew what it was saying.
Still, it's easy to calculate:
1. Go to BEA and get the National output by industry tables
2. Adjust for inflation
3. Divide by population
I made a nice chart of this showing real output per capita in Ag and manufacturing 1983-2023.
Agg $1,718 in 1983 to 1,856 in 2023.
Mfg $22,019 in 1983 to $21,531 in 2023.
Generally, the polynomial trend line was concave up for ag (starting and ending high) and concave down for mfg.
* 1. This was a really quick and effort.
* 2. Agriculture had a boom in the mid to late 70s to early 80s and then declined. So that period is a relative high point that should be noted in comparisons.
* 3. I don't want to get lost in the quibble of whether manufacturing is absolutely declining or simply stagnant. Neither is good.
Again, the argument here isn't to say we should subsidize manufacturing, but rather that we shouldn't regulate it out of existence and then claim that, somehow, this represents a triumph of efficiency.
Hey, it's great if you have a servant to cook all your meals. If you are completely unable to cook your own meals, you might end up being the servant. Are we at that point? No. But we should have the honest discussion instead of crowing about how being unable to cook is efficient.
Manufacturing output and productivity have both grown, but not as fast as those of other sectors in the economy - including sectors that didn’t exist in 1980.
Unless the Chinese, for example, are givng their goods away, then we Americans are paying for the goods with something of equal value to the Chinese (or any other foreign country). I am not arguing that the markets aren't clearing themselves but asking with what are we paying?
Consider two men, Joe and myself. Joe raises cattle and vegetables some of which I purchase each month by selling Joe 0.1% title to my home each of those months. I run a goods and service monetary deficit with Joe but it is all ok because I run a capital surplus with Joe because he is investing in my house?
It's all ok if it's a voluntary transaction. I don't have any interest in judging whether you think you've been ripped off or if you think the government should have prevented you from making a bad deal.
This is one of the cleanest explanations of the trade deficit I’ve seen in a while. I’d add that if the U.S. is going to run persistent deficits, it's at least better that it’s matched by foreign demand for our assets—especially productive ones. The real concern isn’t the deficit itself, but what kind of capital inflow we’re getting in return. Are we selling claims on future income or selling off the farm? That distinction gets lost in the political noise.
If Canada has 200+% tariffs on US dairy, so as to protect its higher cost dairy producers, that is clear evidence of Canada “taking advantage of us”. Not a myth. The existence of many tariffs by other countries on US goods, with far less Expert Economist Disapproval, including Kling & Krugman & Sumner, implies a double standard.
It might be that the sum of benefits & costs of these foreign tariffs are actually a bigger boost to US investors, the rich, and a small reduction in prices to all consumers, so a total positive that helps more than it hurts the US producers who sell less. Maybe net US positive in USD, but the economists are not being honest about how the US workers & producers are losing millions while rich global US investors are making millions + 1% or so.
“In terms of employment … agriculture is even more ‘devastated’ …”
I understand the implication of this sentence.
And yet - a can of peaches at the grocery store - that most basic staple - is now nearly $3 at HEB.
I guess productivity increases can mean fewer acres devoted to peaches, lower yield overall in most areas, but greater productivity as more of it is processed in novel ways, perhaps.
And yet, what I see is that the US population increased by 60 million in the last 25 years (assume then 70 to 80 million, it will always be undercounted now) - and we all get fewer peaches.
Mr. Kling: Is this your Law or someone else? I want to use it for my April roundup of quotes for Entrepreneurs.
"Law of Relative Productivity Growth: In the medium to long run, labor moves from sectors where productivity is rising faster than demand to where demand is rising faster than productivity."
“Law of Relative Productivity Growth: In the medium to long run, labor moves from sectors where productivity is rising faster than demand to where demand is rising faster than productivity.” Arnold Kling in “Trade Myths that Refuse to Die”
Kling observed that this follows from Baumol’s Cost Disease.
I was aware of Baumol's Cost Disease but look at that as a very different insight: certain tasks are amenable to expertise acquired over a decade or two but not as susceptible to automation so their costs go up: e.g. symphony musicians. But I can see it's the negative framing of what you are calling "Law of Relative Productivity Growth:" Thanks!
The real issue is that many people don’t understand anything about international commerce and balance of payments (starting with the US President and most of his advisors). The media also contribute to the bogus narrative by talking trade deficit….if they simply started with a capital surplus the music would sound completely different.
The realer real issue is that it doesn't matter if people understand international commerce and balance of payments. They make trades for their own reasons.
Who cares. If the Chinese want to invest in the US and are prepared to consume less than they produce to procure the usd let it be it. Take a 101 course in economics
That is what I call a very helpful and useful piece of writing. Thanks much.
Seen vs unseen. If one state raises its manufacturing "total labor burden" (minimum wages, liability, taxes, workman's comp, unemployment, regulations, etc.) it effectively outlaws low-productivity and the statistics will show an increase in local manufacturing productivity. There are plenty of scenarios in which, over time, total local output still increases as the population and overall growth increase, but total local consumption of that sector's goods increases far more, helped by lower-productivity imports, especially if those are paid for by selling debt and assets to the makers of those imports, and which can also pay for putting the formerly-above-zero-marginal-product labor force on various kinds of subsidies and welfare since the law makes them below-ZMP now. Hooray!
swing and a miss here, Houseman showed that these MFG stats are hugely biased by the difficulties in accounting for computers. https://conversableeconomist.blogspot.com/2012/12/cautionary-details-on-us-manufacturing.html. She goes on to show the consequence is that trade really is responsible for mfg decline in employment: https://research.upjohn.org/up_workingpapers/287/
It's a bit of a strange argument. She says that if a domestic manufacturer produces by importing cheap intermediate imports, that should count as a trade change, not a productivity increase. But it is both.
Hm. To take Houseman's example to an extreme for illustrative purposes, suppose an American car manufacturer firm moves all its production except for final assembly (screwing on wheels and bumpers) to Mexico, and gets rid of most of its local workforce. (Such 'screwdriver factories' are not unknown, for instance in ex-USSR.) The set of intermediate inputs - the "car kit" - for the final product has become much cheaper, but I don't see how it is reasonable to treat this as a productivity increase by the manufacturer firm in any meaningful sense. It's the same factories, same equipment and work processes, just lower labor costs - lower for reasons totally unrelated to the manufacturer's business. To take the example to the opposite extreme, even if the manufacturer switches the supplier of a basic input such as steel to a cheaper one (foreign or not), why should that count as an increase in productivity? I can see that the manufacturer has discovered valuable information, and that might count towards a productivity increase, but it doesn't feel right. Isn't switching suppliers a red herring? If the manufacturer's existing supplier implemented a new process - increasing _its_ productivity - and reduced prices, the effect on the manufacturer would be the same (lower input costs) as if it had found a cheaper supplier. It is strange to consider the former as a manufacturer's productivity increase, and that makes the latter suspect also. It sort of double-counts the supplier's productivity increase.
Perhaps this is all just terminological confusion on my part, though, using layman's meaning of "productivity" rather than as a term of art in economics. But if the term of art is so far out of whack with the lay meaning of the word, that's hardly good for discussion in this area.
The dramatic increase in the world economy's technological complexity made the advantages of "specialization and trade" even more valuable than in history. This specialization has expanded to the point where a small business can have an international monopoly on a product that every country or customer can make themselves. Free trade allows all possible customers worldwide to take advantage of the lower production costs.
I had a market for "staged frozen zebrafish embryos" with ± 10 minutes of fertilization time, which allowed biological researchers to determine what genes are turned on and off during critical development stages. For example, on neural tube closure, we know which genes are required at what times to prevent spinal bifida from occurring. Remember, we are a vertebrate like zebrafish. I had zebrafish production tanks where I could get tens of thousands of eggs in a few minutes, whereas the research labs with only a few fish would have to spawn them repeatedly to obtain the samples. At a time when the minimum sample size required a thousand or so embryos, I had a market from the Max Planck Institute in Germany to the University of Singapore and every major research university in between. Total sales were not that large, but the amount of Ph.D. level research time saved was huge. My infrastructure cost was covered by my 10+ million zebra fish/yr sales, and Google ads allowed highly targeted research customers to find me.
Science and technology have grown in my lifetime from the time when a "Renaissance Man" who knew at least the boundary of all human knowledge about the world could exist—and I met some of them—to now, knowing the boundary of human knowledge in even a few narrow areas is a stretch. This changes the world, but also allows fools with good rhetoric to make foolish statements in all areas of real knowledge, from economics to biology, and even engineering. You can even get the Surgeon General saying N95 masks don't work when working-class people, where OSHA has been mandating N95 masks for half a century, knew they did.
A lot of the concern is with manufacturing employment: the good jobs problem. There is also a resilience issue: e.g. shipbuilding. The hegemon of the maritime order should have more ship-building capacity. Yes, the Jones Act is an issue, but efficiency uber alles is not the basis of a resilient polity.
Shipbuilding and steel production are two of the most protected industries in the United States, and are among the industries in the worst shape. Protecting people and businesses from competition rarely makes them perform better.
And a great way to get rid of debilitating support is to cite the resilience point.
Brian Potter had a post on this subject at Construction Physics: https://www.construction-physics.com/p/why-cant-the-us-build-ships
The last time the US had an advantage in shipbuilding was when ships were made of wood.
There are many reasons for the decline in the U.S. shipbuilding industry, including:
- High labor costs due to unions and legislation such as the Bacon-Davis Act.
- High domestic steel prices due to tariffs.
- Low productivity due to make-work union rules.
I agree, but most of the decline happened before WWI, when none of these factors were significant. Poor organization and management seem to be the primary causes.
according the article, foreign labor costs and foreign steel costs actually HAVE been big factors dating back to at least the 1860's.
The third factor is when a small island/maritime nation, which already has low labor costs and low steel costs, decided to absolutely devote itself as a matter of honor to having the best shipbuilding organization and management in the world, too.
When you have all three, you become the premier shipbuilder in the world.
The public discourse will always be about employment, but as economists we should be steel-manning the point about output, not misleading people the discussion.
I think the efficiency vs. resiliency paradigm is a good way to look at it. Does it make economic sense for me and my family to eat out every day? Maybe from an efficiency perspective (though I doubt it). Does it make sense that so many people ALWAYS eat out and literally can't cook? Nope.
There are two primary problems with the resiliency argument that I see:
1: Choosing WHAT we should be resilient to. What do we choose to specialize away from given that we can't know how to do everything. I make my own knives and garden tools, but obviously I am weird because I find it fun. Other people can't replace the fill and flush mechanisms on their toilets; I don't find that fun, but it is worth doing myself. The government stepping in to decide what jobs and industries are necessary for people to do would be a disaster, and if the gov. focused only on defense focused industries (which does seem reasonable) the impact on manufacturing would be miniscule. You just wouldn't see millions of people suddenly employed making ships, I think.
2: Arguing that the government will pick the right balance between efficiency and resiliency seems to ignore everything we know from experience about how government functions. Even assuming the government could do that in some fairly effective way, as opposed to finding a spot way inside the production possibility curve like they normally do, empirically the government does not seem at all interested in resiliency. Consider the many departments whose notional role is to maintain resiliency: the Federal Reserve, FEMA, everyone who is supposed to keep wildfires from gutting central CA, HHS, Department of Transportation... Generally the behavior of government is to shift as many resources away from resiliency and towards fancy new things as possible, because the incentives favor putting your name on fancy new stuff and not paying attention to the boring day to day operations. I would absolutely expect bureaucrats to optimize for favoring their buddies and flashy numbers at the expense of resiliency. When the crisis comes and their resiliency fails, best case for them they will be retired, and even if not, well who could have seen this coming? It is unprecedented!
Just in terms of our current problems, we didn't have to be over a barrel for rare earth magnets. The government absolutely could have made cut outs in the EPA regs for these important defense related items but, nah, why do that? The government protected the mohair industry from outside competition, but not the rare earth mining and refining industry from other branches of their own government. There is a lesson there.
Yeah, to be clear, I'm not arguing for government to come in and pick industries to subsidize.
Rather, I'm pointing out that, I think in large part, our current state of affairs is implicitly based on a lot of government subsidization. What's "efficient" for the market to produce is based on what the government subsidizes (or not) by regulatory policy. With respect to trade, it's "efficient" to import things because they're produced in ways that would be illegal to do domestically.
I still find that kind of "efficiency" reasoning morally repugnant. If slavery is legal in Y, and this makes trading with Y efficient, the answer isn't
a) legalize slavery in the US
b) subsidize US industry
c) make war on Y
It's simply
d) don't trade with Y
Now, in many cases that aren't so morally repugnant as slavery, the answer would be A, reduce the legal and regulatory burden so that what is illegal in the US but legal in Y becomes legal in the US.
More generally, we presume economic efficiency based on voluntary trade. But the greater the divergence in legal rights between trading partners, the less that presumption can hold.
I don’t think you are entirely wrong, but I also think most of our regulation goes well past the horizon of “morally acceptable” and into the territory of “crazy people made demands and idiot politicians followed up for votes.” We should be rolling back lots of regulations and getting government thumbs off the scales as much as possible, not adding a few more thumbs with tariffs.
Just to the point on slavery, slavery is a lot less efficient than people tend to think. Unless there is a huge dearth of workers compared to capital, a state that does not obtain in most of the countries people are worried about, there is not an advantage for slavery.
Just because resilience may be hard does not mean it does not matter. Mainstream economics is getting migration catastrophically wrong, particularly in Europe, because it pays no attention to resilience.
https://www.lorenzofromoz.net/p/economics-a-discipline-committing
I am not saying resilience isn't important or does not matter. I am saying that I will believe government might be capable of choosing the right trade off between efficiency and resilience when they stop talking about resilience while actively doing things that make society less resilient. Getting government, which has both extremely limited ability to understand what is needed for resilience and almost zero incentive to find out, to serve this purpose is rather a fool's errand I am afraid.
As to economics on immigration, economists are getting it wrong because they are assuming away all those pesky details like "institutions and culture matter" and "whoops, we have a welfare state." Not to mention "we hate to actually enforce property laws, especially against brown people." I don't disagree that economists get themselves blinded by ideology now and again; the AEA is a bloody mess. That doesn't mean that one can say "resiliency matters, and I believe this will increase resiliency" and have it be true; it is worth discussing how that would actually happen with reference to institutions as they actually exist.
People will do almost anything for enough money, but free employment stops offering more money when it exceeds the value gained from the work; slavery does too, but the value paid includes guards, overseers, patrols, and other overhead, because people value their freedom. I don't know what economists call this, but to me, it means that slavery can never extract as much work from slaves as it could from free men. The only work which benefits from slavery is that which no free man would do, mostly because it kills its workers, reducing further how much training is worthwhile with slaves.
I don't see slavery as anything but control freakery.
Adam Smith agreed with you.
Even when slaves were really well paid (eg salt miners in the UK) and only couldn't leave, they still had to be kept under lock and key (figuratively speaking in that case).
Your "Choosing WHAT we should be resilient to" is the main problem, to me. It assumes that "we" should choose such things, "we" should steer the economy instead of letting it take its own course.
That "we" can only mean government, and not only does government in general have a poor track record in everything they mismanage, but elections change priorities and policies so often that no choices last long enough to do anything but muddy the water.
It's like a bunch of drunks taking turns trying to steer a raft, as if they had any control over where the river goes. They are more likely to run aground or capsize it.
Companies want to survive, so they have an incentive to consider resiliency. What are the sources of their inputs? What are the threats to those sources? How can we best handle those risks?
Yes, and those are best solved by the companies themselves, not government with its one-size-fits-nobody policies.
But not the resilience of the system as a whole. Biological systems select for it. Eventually, they fail (all species disappear eventually) but lineages continue because resilience is what is selected for. Polities need to concern themselves with resilience.
Mainstream Economics is getting migration catastrophically wrong, particularly in Europe, because it pays no attention to resilience.
https://www.lorenzofromoz.net/p/economics-a-discipline-committing
I don't understand what you're saying. I get the impression you think politics, ie governments, is capable of getting resilience right. Or, IOW, you think governments are smart enough and have enough data to manage economies better.
I don't understand how anyone can believe governments know more than people with skin in the game.
from Alex Tabarrok's "The Prophet's Paradox":
"The political problem of disaster preparedness is especially acute for the most useful form, disaster avoidance. The problem with avoiding a disaster is that success often renders itself invisible. The captain of the Titanic is blamed for hitting the iceberg, but how much credit would he have received for avoiding it? ...
"The prophet’s paradox can undermine public support for proactive measures. The very effectiveness of these interventions creates a perception that they were unnecessary, as the dire outcomes they prevented are never realized. Consequently, policymakers face a challenging dilemma: the better they manage a potential crisis, the more likely it is that the public will perceive their actions as overreactions. Success can paradoxically erode trust and make it more difficult to implement necessary measures in future emergencies. Hence, politicians are paid to deal with emergencies not to avoid them"
One of the tragedies of life.
https://marginalrevolution.com/marginalrevolution/2025/04/the-prophets-paradox.html
Indeed. The optimal state for politicians is for the disaster to happen in such a way that it isn’t obviously their fault, be minor enough that it is easy to fix yet be hyped into the worst thing since the Cambrian extinction, then to be seen solving the problem while mostly just talking about it and spending the political capital elsewhere.
The last thing bureaucrats want to do is solve the problems that keep them employed, and politicians are just bureaucrats with bigger egos.
I know this is meant metaphorically, but I am too literal-minded not to connect it to the news a few years ago that Del Monte was shuttering many of its canning facilities, including one that was important to farmers south of here.
The company said it was going “asset-light”. Since it’s a canning company I’m not sure what will be left … little plastic containers of fruit, I guess, as people eat more like toddlers now perhaps.
We are supposedly all such foodies now!
But the consumer who doesn’t know how to make use of canned goods, thriftily, and tastily as part of home cooking - is a stupid consumer.
And stupid consumers will never be good producers, or much else.
Thank heavens the government seems to have planned for that, with economists in its (academic) pay floating UBI.
Looking at economic aggregates can miss important details. When we no longer have adequate manufacturing capabilities in munitions, pharmaceuticals, semiconductor chips, as well as shipbuilding, we are vulnerable. And don't we owe something besides welfare to our working and middle class fellow citizens whose jobs are affected by imports of finished goods and cheap labor?
Much of America’s munitions are still made in outdated, Army-owned plants. Thankfully, these are finally being replaced with modern facilities. Military procurement has long been a national scandal.
The next major war is likely to center around cyberattacks, cheap drones, and AI-guided missiles — not just tanks and aircraft carriers. Unfortunately, the U.S. military continues to invest heavily in expensive, traditional weapons systems like carriers and manned jets, even though they’re increasingly vulnerable to cheaper, decentralized threats.
Many service-sector jobs — like logistics, software engineering, communications, and cybersecurity — are now essential to modern warfare. War today isn’t just about building tanks and planes; it’s about information, coordination, and adaptability — all areas where service-sector skills are critical.
Our real vulnerability isn’t a shortage of factory workers. It’s the government’s failure to adapt its spending and strategy to how fast warfare is changing.
The U.S. has 149 pharmaceutical manufacturing plants employing about 52,000 people. These plants produce more than half of the medicines consumed in the country. The leading foreign suppliers of medicines, by dollar value, are Ireland, Germany, Switzerland, India, and the Netherlands.
Less than 8% of America’s 166 million workers are employed in manufacturing today. Studies show that every manufacturing job “saved” by tariffs or trade barriers ends up costing about six or seven jobs elsewhere in the economy. Don’t we owe our working- and middle-class fellow citizens something better than protectionism, high prices, high unemployment, and welfare?
"by dollar value"
This is due to patents.
Most of your generic drugs, that amoxicillin you used in your last infection, come from China/India and a lot of the raw ingredients India uses come from China.
In fact faced with the prospect of sanctions the cost of generic drugs could be going up quite a bit. We can pay it, but of course in a war if wouldn't be a 100% markup but an infinity markup.
As I said, there are tradeoffs to be made. I don't think either an absolute protectionist or open position is great. It is not a matter of finding the one right solution with no downside, which doesn't exist.
I'll rephrase that for the flip side.
Don't we owe something besides higher taxes and more expensive goods to our working and middle class fellow citizens whose purchases are affected by bans on cheap goods and better jobs?
I am by no means an enthusiast for egalitarianism, but I think getting us cheap goods needs to be balanced against the more serious social problem of un- or under-employed fellow citizens.
Balanced? By whom? Oh yes, the government, by busybodies who know more than the individuals concerned, and who change policies every few years for no reason other than to distance themselves from the predecessors.
The government is unfortunately going to arrange trade in any event; the question is what factors will influence their decisions.
Whatever factors do, they won't be the right factors, they will be political factors which are irrelevant or wrong for almost everybody but a few favored cronies.
Us arguing about what government should do is meaningless in any practical sense. The only sense left is the idealist sense.
I am a nobody. Let me reference a somebody . In a Fortune 2004 article Warren Buffett advocated for tariffs /a coupon system to deal with the trade deficit. He likened the situation as two islands where one island gradually is in debt and then ultimately gets owned by the other island. An interesting read and expands the discussion. Treasuries are one thing and hard assets are another.
I beleive economists have become too comfortable with their narratives on myth 1 and are glossing over important things they would highlight in other cases.
Output, rather than the number of workers is the appropriate measure of myth 1. In real dollars, manufacturing output has stagnated in the US.
You point out that manufacturing output has doubled since 1980. But that tells us nothing about the relative productivity of mfg. The economy has grown 8-10x in that period depending on whether you are counting for inflation.
To draw a simple analogy, although farming employment has fallen, we are more productive and better off because we can grow everything we grew in 1980 and more. A simple way to say it is agricultural GDP per capita has risen.
Manufacturing GDP per capita has fallen though. We literally cant make what we used to make.
I'd like to see some sources for those claims. Everyone else I've seen writing about this shows the opposite, that both manufacturing output and productivity have increased and are still increasing, although the COVID lockdowns scrambled the data for a while.
To me, it's telling that this data isn't readily available. It's kind of like with a lot of crime and immigration statistics. I don't think a lot of research and looking is done, because the results won't win you anything.
At first I just asked Google AI, but when I looked closely, it wasn't 100% clear the AI knew what it was saying.
Still, it's easy to calculate:
1. Go to BEA and get the National output by industry tables
2. Adjust for inflation
3. Divide by population
I made a nice chart of this showing real output per capita in Ag and manufacturing 1983-2023.
Agg $1,718 in 1983 to 1,856 in 2023.
Mfg $22,019 in 1983 to $21,531 in 2023.
Generally, the polynomial trend line was concave up for ag (starting and ending high) and concave down for mfg.
* 1. This was a really quick and effort.
* 2. Agriculture had a boom in the mid to late 70s to early 80s and then declined. So that period is a relative high point that should be noted in comparisons.
* 3. I don't want to get lost in the quibble of whether manufacturing is absolutely declining or simply stagnant. Neither is good.
Again, the argument here isn't to say we should subsidize manufacturing, but rather that we shouldn't regulate it out of existence and then claim that, somehow, this represents a triumph of efficiency.
Hey, it's great if you have a servant to cook all your meals. If you are completely unable to cook your own meals, you might end up being the servant. Are we at that point? No. But we should have the honest discussion instead of crowing about how being unable to cook is efficient.
Manufacturing output and productivity have both grown, but not as fast as those of other sectors in the economy - including sectors that didn’t exist in 1980.
With what, precisely, are we, Americans, paying for the excess goods we are consuming?
What excess goods? If goods are bought, they self-evidently aren't excess. Something about markets clearing themselves, but I am no economist.
Unless the Chinese, for example, are givng their goods away, then we Americans are paying for the goods with something of equal value to the Chinese (or any other foreign country). I am not arguing that the markets aren't clearing themselves but asking with what are we paying?
And just to illustrate my reason for asking:
Consider two men, Joe and myself. Joe raises cattle and vegetables some of which I purchase each month by selling Joe 0.1% title to my home each of those months. I run a goods and service monetary deficit with Joe but it is all ok because I run a capital surplus with Joe because he is investing in my house?
It's all ok if it's a voluntary transaction. I don't have any interest in judging whether you think you've been ripped off or if you think the government should have prevented you from making a bad deal.
But is it smart?
Smart for who? Judged by whom?
So, you have no opinion on whether or not it is smart to sell yourself into debt to someone else?
This is one of the cleanest explanations of the trade deficit I’ve seen in a while. I’d add that if the U.S. is going to run persistent deficits, it's at least better that it’s matched by foreign demand for our assets—especially productive ones. The real concern isn’t the deficit itself, but what kind of capital inflow we’re getting in return. Are we selling claims on future income or selling off the farm? That distinction gets lost in the political noise.
If Canada has 200+% tariffs on US dairy, so as to protect its higher cost dairy producers, that is clear evidence of Canada “taking advantage of us”. Not a myth. The existence of many tariffs by other countries on US goods, with far less Expert Economist Disapproval, including Kling & Krugman & Sumner, implies a double standard.
It might be that the sum of benefits & costs of these foreign tariffs are actually a bigger boost to US investors, the rich, and a small reduction in prices to all consumers, so a total positive that helps more than it hurts the US producers who sell less. Maybe net US positive in USD, but the economists are not being honest about how the US workers & producers are losing millions while rich global US investors are making millions + 1% or so.
“In terms of employment … agriculture is even more ‘devastated’ …”
I understand the implication of this sentence.
And yet - a can of peaches at the grocery store - that most basic staple - is now nearly $3 at HEB.
I guess productivity increases can mean fewer acres devoted to peaches, lower yield overall in most areas, but greater productivity as more of it is processed in novel ways, perhaps.
And yet, what I see is that the US population increased by 60 million in the last 25 years (assume then 70 to 80 million, it will always be undercounted now) - and we all get fewer peaches.
No, we didn’t starve.
Mr. Kling: Is this your Law or someone else? I want to use it for my April roundup of quotes for Entrepreneurs.
"Law of Relative Productivity Growth: In the medium to long run, labor moves from sectors where productivity is rising faster than demand to where demand is rising faster than productivity."
I may be the one who calls it a law, but it is very closely related to what goes by the fancy name of Baumol's Cost Disease.
I included this in my April post https://www.skmurphy.com/blog/2025/04/30/quotes-for-entrepreneurs-curated-in-april-2025/
“Law of Relative Productivity Growth: In the medium to long run, labor moves from sectors where productivity is rising faster than demand to where demand is rising faster than productivity.” Arnold Kling in “Trade Myths that Refuse to Die”
Kling observed that this follows from Baumol’s Cost Disease.
I was aware of Baumol's Cost Disease but look at that as a very different insight: certain tasks are amenable to expertise acquired over a decade or two but not as susceptible to automation so their costs go up: e.g. symphony musicians. But I can see it's the negative framing of what you are calling "Law of Relative Productivity Growth:" Thanks!
there is an area of the USA called the rust belt
there is no area of the USA called the fallow plains
None of my business. It's your deal.
The real issue is that many people don’t understand anything about international commerce and balance of payments (starting with the US President and most of his advisors). The media also contribute to the bogus narrative by talking trade deficit….if they simply started with a capital surplus the music would sound completely different.
The realer real issue is that it doesn't matter if people understand international commerce and balance of payments. They make trades for their own reasons.
Absolutely. The problem is political interference. Would be great to live in a tariffs free world.
Tell that to the Chinese.
Who cares. If the Chinese want to invest in the US and are prepared to consume less than they produce to procure the usd let it be it. Take a 101 course in economics
Why? Why is it anyone else's business what the Chinese government does inside China?
If you think it is anyone else's business, then you concede it is China's business what the American government does inside America.
I doubt you think that.
A quintessential application of Occam's Razor! Thanks for this cogent review of the logic.