60 Comments

Another fascinating "five things."

How concerned are you that industrial policy will fall prey to its usual predators––graft and inefficiency and special interests?

As for the working class, the book "Second Class" is an interesting read about the attitudes of the working class. I recommend it.

Expand full comment
RemovedMay 27·edited May 27
Comment removed
Expand full comment
May 28·edited May 28

Nobody ever defines what constitutes the "working class." In my experience, a lot of what you said there applies to small business owners who are in now way the "working class" but nonetheless feel a lot of what is described here.

Expand full comment

If America wants to create globally competitive manufacturing, it can't do it alone. German manufacturing would not have survived without outsourcing a bunch of their work to southern and eastern Europe. If you want America's supply chains to compete with Asia's you have to integrate central America and eventually the rest of Latin America into NAFTA.

Expand full comment
RemovedMay 27·edited May 27
Comment removed
Expand full comment

That's insane. A wealthy Mexico would be a stable partner and reliable ally. Russia can't tolerate wealthy neighbors because it believes that those neighbors are meant to be their slaves. Unless you're offended by the idea of a nearby nation self-determining, a stable, healthy Mexico would be a MASSIVE benefit to not only the U.S. but the entire Central American region.

Expand full comment
RemovedMay 27·edited May 27
Comment removed
Expand full comment

Gotcha. So we cannot allow any other nations to be wealthy or successful, check. Certainly, our constant military struggles with Canada and the EU demonstrate that!

I try to be polite here, so there's really nothing left that I can say to you. Seek educational help.

Expand full comment

Mexico would first need to root out and defeat thoroughly its drug cartels. The problem is that the cartels have state capacity in violence, logistics, infrastructure and economics that they are breakaway nations within its own borders. Also, the U.S.-Mexico border cities are tightly integrated economically and culturally and don't want or need war to achieve their goals.

Expand full comment
May 27·edited May 27

Intersting take, thanks.

Then again, we’ve survived with a wealthy Canada on our borders. And I think it would be better to have a prosperous Mexico than a corrupt narco state. Countries like Colombia and Peru have made large strides over several decades in both reducing domestic disorder and improving economic conditions without becoming bigger threats to their neighbors. Countries like Chile and Costa Rica and Uruguay have been stable for a long time - would love to see Mexico become Uruguay or Chile, though that may not be possible for larger, diverse states with high inequality like Mexico and Brazil.

Fully agree with Dr. Ahmed that we need to leverage labor and materials resources in our hemisphere (we do this infinitely better in Latam than Europe does with Africa) though I think with increasing automation, AI and robotics, manufacturing in the future may well depend more on land prices, electricity prices and availability, transportation network (freight rail, highways, ports) than on cheap labor. In these things the US is strong.

What we lack is a machine tool industry (and much of a robotics industry) to build machinery necessary for manufacturing. Germany, Japan and S Korea long ago ate our lunch. Overall, there are a lot of things I’d see government invest in (including basic research) before I’d settle on paying rich people to buy EVs with Chinese battery IP as our prime industrial policy!

Expand full comment
RemovedMay 27·edited May 27
Comment removed
Expand full comment

Since democracies don't go to war with each other, the US should be concerned about Mexico's form of government, not how wealthy it might become.

Also, Mexico's geography alone mitigates it outstripping US economically.

Expand full comment
RemovedMay 27·edited May 27
Comment removed
Expand full comment

Whether any democracies have gone to war with each other depends upon the definition of democracy and war.

If we consider only principal combatants (such as in a hypothetical 21st century Mexican-American war), this then excludes examples like Finland's alignment with the Axis powers. If we only consider states with at least majority suffrage, civilian control over militaries and moderate to high scores on democratic spectra, then there are very few examples, all various forms of wars stemming from independence or founding, and ongoing decolonization disputes. Examples include the First Balkan War, various Indo-Pakistani conflicts, wars between Israel and Lebanon before 1975.

So virtually every example comes with an asterisk or three.

I therefore have qualified confidence in my statement, but admit it requires a number of qualifiers, that others might not agree are germane.

Now, if we relax the requirement for majority suffrage (so women's suffrage is not required), interestingly, we get candidates like the Philipine-American War, and the Mexican-American War. So maybe you're right that we shouldn't discount future aggression on the US's southern border. That said, I think war more likely to occur with a weak de facto narco-state in northern Mexico, rather than a prosperous Mexico with robust democratic institutions able to impose order on its boundary states.

Expand full comment

There are lots of ways China is a potential threat.

Expand full comment

Wouldn’t a wealthy Mexico greatly *improve* the security situation? Human trafficking (particularly including false promises of immigration and asylum) and the drug trade are the two big security threats from Mexico, and they would both be greatly mitigated by an improvement in wealth. I don’t understand what the new threat would supposedly be if there was a triple sized Canada.

Expand full comment
RemovedMay 27·edited May 27
Comment removed
Expand full comment

I think you’re making some leaps here. Does a wealthy France make the US security situation worse? I can see that it diminishes US ability to make unilateral international decisions, but that doesn’t diminish security. Wealthy Canada similarly increases security while decreasing US ability to do what it wants unilaterally. As far as I can tell, greater wealth for anyone generally means greater security for everyone, unless there are specific conflicts the people have. Germany and France have done well for each other over the past 70 years of wealth.

Expand full comment

The aftermath of World War II did also produce an internationalist consensus that was determined to avoid an outbreak of war as devastating as WWI and WWII.

WWI marked the beginning of industrial-scale warfare, which ratcheted up troop casualties as well as destruction of land and buildings. I think Piketty in "Capital in the 21st Century" argued that the world wars wiped out several centuries of economic gains before it. The three-decade postwar rebuilding period led to an unprecedented rise in economic growth and material living standards for people that the postwar era was probably the most successful land-reform regime in history.

Expand full comment
RemovedMay 27·edited May 27
Comment removed
Expand full comment

Desire modification - a life-saver for the North Korean state, and totalitarian regimes generally!!

Expand full comment

https://www.nytimes.com/interactive/2024/05/07/climate/battery-electricity-solar-california-texas.html

Corn ethanol doesn’t pencil. Full stop. If you replace the acreage used to grow corn for ethanol with solar panels, you would generate almost three times the electricity needs for the U.S. And you can grow crops under solar panels, which saves significant amounts of fresh water lost to evaporation. Also, greenery under solar panels reduces heat and makes a solar panel up to 20% more effective at generating electricity.

There is a suite of solutions to AltGreen Energy. The federal funding is in place. We simply lack the vision and political will, so far.

Expand full comment

We should stop growing corn for ethanol, but it would be a shame to cover Iowa and Nebraska with solar panels. We could just use it for food, after all tomorrow is Taco Tuesday, so let’s make some tortillas.

Expand full comment

Tortilla technology has really advanced, due to eaters having allergies to the primary grains of corn and wheat (flour). Some other tortilla bases: chickpea, tapioca, rice, almond, coconut or teff.

Expand full comment

The modern peer review apparatus was basically invented in the 1960s

https://www.experimental-history.com/p/the-rise-and-fall-of-peer-review

Fortunately, the scientific method has always had a built-in system for dealing with the problem of fakery - it's called independent replication of experiments.

When I decided to diagram how the modern scientific method works, in service of this post:

https://cbuck.substack.com/p/extraordinary-evidence-requires-extraordinary

it took me quite awhile to realize that the diagram really does *not* need a peer review box. It just needs the orange "independent validation" box that has been there ever since Galileo. The orange box seems slow and inefficient, but I'm with Mastroianni - we did the peer review experiment for half a century and the conclusive result is that it really isn't an adequate substitute for independent verification. The best available solution to the fake data crisis is a simple cultural shift back toward respecting the importance of the orange box.

Expand full comment

In terms of civilian impact, the most important airplane of World War II was the Douglas DC-3, a cargo plane. About 10,000 of them were built in the United States during WW II. After the war, they continued to be used as cargo planes and were also converted for passenger service, holding 21 to 32 passengers. Obviously, the US Air Force had a lot of them to get rid of, and this helped greatly to jumpstart commercial aviation.

The DC-3 was regarded as invulnerable. A friend of mine said it incorporated about twice as much aluminum as would be used today, so that it was somewhat of a flying tank. There are still 89 DC-3s registered as flying in the United States today. Many of these vintage DC-3s have spent more than 10 years aloft.

I remember reading about a small regional airport in the Great Lakes that used a DC-3 depending on the number of passengers booking a flight, but most DC-3s are used for charters or at air shows.

One should realize, their electronics and other systems have been greatly updated, but the air frame was built for the ages.

Expand full comment

I don't think fake science is really on the rise. I think it's been largely fake for decades. I base this opinion on having read a lot of really visibly bad science papers, especially in the past few years when I did a deep dive into COVID (but have read papers on other topics too, including those in my direct area of expertise). What's new is open access + the internet leading to a wider awareness of how just how brazen fraud in academia really is, but if you want you can easily find garbage studies from the 1960s. This is a big part of the replication crisis in psychology: well known results taught in schools for decades, like Zimbardo, turned out much later to be built on lies.

As you observe, the problem is culture. Academia is run by weaklings who shy away from conflict and will only fire someone for ideological reasons, as far as anyone can tell. Therefore there's just no point in reporting fraud because there won't be any outcome. At best, with a lot of effort and by making a lot of enemies, you might eventually get a retraction (nobody cares). What you won't get is the perp being fired or prosecuted, or new systems being put in place to stop similar future frauds. You'll also likely end your own career and maybe get a lawsuit for your troubles.

This leads to a downward spiral in which students and young academics become numb to the bad behavior, come to accept it as just the way things are, then eventually become corrupted by it because fraud is the winning move if you want a career (and many of them have specialized in fields employers don't care about so they're trapped in academia). Finally they become the teachers of the next generation and then pass on the fraudulent techniques as "the scientific method". This results in a perpetual motion machine of junk that's impervious to reform, because the people doing it have so long ago convinced themselves they're the good guys.

You say, maybe people should be rewarded for quality. Great idea, except the system is already trying to approximate that. There's no direct way to measure quality because the word "quality" doesn't mean anything specific, so it devolves to measuring quantitative proxies for quality like citation counts. This is nice but breaks down when entire fields are corrupt, because they all cite each other without producing real science. Climatology is a great example of a field that is nearly pure fraud but whose participants have managed to repeat their lies so often that they're now official truth, with anyone who does tell the actual truth being punished as dangerous liars. Goebbels really had nothing on these guys!

There's no fix for any of this. I searched extensively. Every time I came up with a bright new idea for how to reform the system I'd do some checking and discover it was already tried but academics immediately worked around it. Eventually it became clear that it's impossible to reform a system in which almost everyone is corrupt. Some people, somewhere, must care and must be willing to fight for what's right, or else any reform attempt is doomed. Moreover they must have some power. Universities do not seem to have those people no matter how high you go. Without any chance of reform the whole thing is just unfixably toxic. The only fix that can ever work is if governments conclude they don't know how to fund science and just giving up (i.e. defund public sector science research entirely).

Expand full comment

This is not at all my experience in the field of computer science. Yes, paper counting is a problem, yes people are publishing too many incremental results, but no: there is no epidemic of fraud. Statements like “everyone is corrupt” have no relationship with what I see, which is a huge number of excellent results by hardworking scientists who could all make way more money switching to industry, but stay in academia because they love science.

Moreover anytime I see someone conflate the replication crisis (people doing poor studies and using lousy statistical techniques on low-effect-size experiments) with outright fraud (like faking imagery) I know that we’re dealing with someone who doesn’t really understand the issues, or has an axe to grind against science in general. It feels from your post that you have a very big axe to grind.

And of course you couldn’t help but include the requisite dose of climate denialism, which I strongly suspect is your real issue. The good news is you don’t need scientists anymore: temperatures have increased enough that you can just go measure them yourself.

Expand full comment

If you only read the CS literature then your reaction is totally understandable, and I'd have the same reaction to my post. What you don't realize is how much standards vary between fields. Invest some serious time into the literature of other fields and compare the level of honesty with the output of CS departments. They're from different planets.

The replication crisis is not exclusively or even primarily caused by lousy statistics. Failure to replicate can be caused by many different things, including, most obviously, that the data you're trying to replicate was fake to begin with. You seem to think the two are totally disconnected but that's not the case. People originally focused on poor study design back in 2011 when the replication crisis first started picking up steam, because that problem can be spotted easily just by reading papers. A paper that claims to use a solid methodology but in which the experiments never actually happened isn't going to replicate, but it won't be immediately clear why not. Figuring out that it's fraudulent takes a very different set of skills to just spotting that the sample size is 30 undergrads. So, of course people focused on the low hanging fruits first.

Lack of replicability is not really the biggest problem. It's papers that replicate but which are using invalid methodologies, or which are just dishonest. I see this confusion a lot, where people use "it can be replicated" as a synonym for "it's a high quality and true claim". There are lots of ways for a replicable paper to be dishonest, like by making a claim in the abstract that isn't supported by the collected data. You never see this in computer science papers, but you see it all the time in many other fields.

I don't care about climatology specifically. There are lots of bad fields. Climatology just happens to be one of the worst, methodologically speaking. Nothing in climatology can be replicated. It's funny you say go measure the temperatures yourself. You'd think that'd work, right? Try it! Please do try it. You'll discover you can't replicate the size of the claimed warming or even get close to it.

It's bizarre, but nothing in climatology can ever be said to replicate, even though it's all supposedly based on public observational data. The reason is fraud: they've developed a habit of changing the historical data any time it stops agreeing with their theory. Compare the IPCC report in 2013 which talked about a long term "pause" or "hiatus" in warming. Recorded temperatures hadn't really gone up for 15 years and it was a big mystery why. Thousands of papers were published on it, here's one example but search Google Scholar for global warming pause and you'll find many more:

https://www.nature.com/articles/nclimate2341

What caused the pause? In the end we have no idea because climatologists found a super "scientific" fix for the mismatch between data and theory: they just changed the historical data.

https://www.theguardian.com/environment/2015/jun/04/global-warming-hasnt-paused-study-finds

Climatologists and the IPCC were studying the problem in 2013. By 2015 we're being told that the "Reassessment of historical data and methodology by US research body debunks ‘hiatus’ hypothesis used by sceptics to undermine climate science". A clear pattern observable in official temperature data, studied by climatologists themselves, was smoothly rewritten as a "hypothesis" by "sceptics" that was used to "undermine" the science. This immediately invalidated 15 years of the prior climatological research literature, but of course none of those papers have ever been retracted.

In reality what this shows us is that the temperature data is meaningless. That wasn't the first time they edited the historical data to make it fit their claims. Any time it stops making their field look important, they change the algorithms they use to post-process all the data, and magically the warming comes back. You can't actually say anything about temperature trends, because tomorrow they might change history again and now your previously true statement is suddenly false. Terms like science, replication, finding, hypothesis don't make any sense in such a field. It looks scientific on the surface, but the core tenets of science have been lost.

Expand full comment

1) "I don't care about climatology particularly"

2) Proceeds to write several paragraphs about climatology, complete with detailed links.

Of course you care about climatology. Back in the real world, Pakistan just crossed 52 degrees C and this is becoming an increasingly common occurrence. I wish you well in the future we've built ourselves.

Expand full comment

I can post links for several different fields. The post you replied to started by criticizing COVID researchers, but I guess you didn't feel like jumping to their defense. A good idea. You picked on climate "denialism" specifically so you got links about why it's rational to deny claims by climatologists. Right?

To repeat myself, announcements about record breaking temperatures are meaningless. You have no idea what the temperature trend is in Pakistan. Climatologists and media routinely do things like announce record breaking temperatures that were recorded for 30 seconds at weather stations directly behind runways, at the exact same moment military jets are landing (and then they lie about it). They report record breaking temperatures that are lower than their own previously announced records (because they change the past it's possible for records to go down as well as up). They report record breaking temperatures that differ by 0.05 degrees C based on sites that have confidence intervals of 2-5 degrees C. They report record breaking temperatures from weather stations that have recently had air conditioning exhaust vents placed right next to them, and ignore it.

You're a mathematically minded man. If I'm really so wrong about the state of the science, there'd be no harm in evaluating their methods for yourself.

Expand full comment
May 27·edited May 27

Excellent! And that was only five of ten, wow.

Thanks for your hard work and also the shout out to the Medicare study. Hospital costs are a huge problem in the US and more out of whack with what the rest of the world charges than most pharmaceutical prices are.

Expand full comment
May 27·edited May 27

Something that's always oddly missing from your analyses is *graft*. Our permitting and regulation system exists in this atrocious state because it massively profits the permit-issuers and regulators. Likewise our industrial policies are based almost entirely around pouring money into the pockets of the policymakers via the companies they own. Our system isn't equipped to build or produce anything; it has been repurposed *entirely* as a system for funneling money, much of it collected from the taxpayer, into the pockets of the upper class, who don't pay taxes. A "stealth feudalism". You're always calling to repair the frame of the house without ever discussing the termite infestation that's eating it in the first place.

Expand full comment

I’m originally from Nebraska, with one side of the family working as farmers. Farmers are conservative until you prove to them there’s a better way to farm. You offer a farmer lease money to have a wind turbine on his land, he/she will take it. You offer a farmer lease money to put solar panels on his land, which will garner him significant savings in water lost to transpiration, he/she will take it. Yields will be better on both solar panels and crops. On a hot summer day, as much as 75% of water via pivot irrigation is lost to transpiration and ends up in the ocean. Given fresh-water shortages, record heatwaves, and record extended droughts, how long can we keep doing the same wasteful things. Much of farming is corporate farming. It’s a business, which is why farmers grow corn to put in your car. And much of the corn that does enter the processed food industry ends up as corn syrup additives, which is causing an epidemic of obesity, diabetes, and cardio-pulmonary disease. The health costs are astronomical, and taxpayers are subsidizing the source. Why not grow soybeans? There are Nebraska farmers who have sold soybeans to China, farmers who have decades-long business relationships with the Chinese. When Trump put certain tariffs in place, he damaged those business relationships. Farmers were not happy.

Expand full comment

My dad served 36 months of combat during WWII in the U.S. Army 1st Armored Division — across North Africa to Sicily to Italy to Germany. Wounded, he recovered in a field hospital and refused to come home. Three brothers went, two came back alive. One left college (mechanical engineering) to join the U.S. Army and served as far as North Africa. Another Uncle left Medical School to serve in U.S. Army Field Hospitals, treating wounded. His sister, my mother, worked in the White House, managing FDR’s correspondence. Her sister, my Aunt, was a Rosie the Riveter in Southern California. There were more. This is not to brag. You can throw a rock and hit a descendant of a family that was totally engaged at this scale in the WWII effort. Everybody did their best.

Of course, there were a few people like Ronald Reagan who claimed “poor eyesight” and remained in Hollywood making Army training films. Gosh, how was he able to read cue cards during his movie career? Other Hollywood stars went off to WWII, e.g., Jimmy Stewart flew bombing missions in the U.S. Air Force.

If I remember correctly, Congress passed legislation stating that any white papers which received government funding for the associated research must be published open source. I wonder if this will help or exacerbate the problem of published scientific fraud? There will be more eyes on white papers because few can afford the high prices of many journals.

Expand full comment

Interesting as usual. As an aside, I really wish the term “Neoliberal” could be retired. The system that we currently have in place is not the result of some sordid consensus between democrats and republicans based on some magical faith in free markets. Democrats have always wanted more taxes, more social programs, more redistribution, and more regulation. We’ve ended up where we are because of political deadlock. The two exceptions, I’d say, would be trade and financial regulation, where there really was a “neoliberal”, consensus. But very few people actually mean that when they use the term.

Expand full comment

The negotiations for NAFTA also had bipartisan support, except from Midwest Democrats rightly worried about deindustrialization. The popular imagination has NAFTA as Bill Clinton's baby, but the negotiations began in the final years of Reagan's term and throughout Bush I's term.

Expand full comment

"Deregulation" - maybe not what I would think of compared to what Noah says. Deciding to actually focus on improving things - whether regulations, efficiency, etc. is a Very Good Idea - generally lost today in the culture war. There are great regulations that prevent a ton of externalities that we all pay for - as well as stupid regulations that are poorly written or don't allow for smarter solutions to be tried......

Ideally (IMO), the goal of regulations (at least related to economy) should be a set of "rules" that govern the game (think about basketball games with too many rules to no rules).... Try to set the boundaries without dictating the "solution". In Practice, this is often hard to do which is why sometimes the regulations have to be heavy-handed, but yes, by all means look at proposals for improvements (also look at implementation efficiency, but after working in large companies, I am afraid that may just go with being big - unfortunately).

Expand full comment

Usually "deregulation" is just the buzz word for pushing off externalities on to other people in order to make bigger profits....

Expand full comment

There are probably some limits to how much we can modify desires in people. It's one thing to make people feel more or less hungry or thirsty or horny, but put someone into a desire modification machine to change their favorite type of music from rap to classical, and I'm pretty sure the person walking out of the machine is not the same person who walked into it. I would fully expect the machine to kill me - overwriting the operating system of my brain to no longer be me. My choice of music is emergent from aspects of my personality and almost certainly can't be isolated and replaced without changing who I am.

Expand full comment

Semaglutide and GLP1 doesn’t actually act on the desire parts of the brain, it instead binds to the satiety parts. So it doesn’t decrease your desire for your first beer or a Rolling Stones concert, but does cut down on how much you want your third beer or second encore. It also doesn’t make you want broccoli instead of Twinkies but obesity isn’t caused by an occasional treat, but by too many of them.

Expand full comment

"that a smart rewrite of the act would be able to protect air quality just as much while not blocking factories, power plants, or housing. "

In the abstract, certainly true. In a practical current political world - absolutely impossible.

Expand full comment

The only argument for industrial policy and general protectionism that resonates with me is scaling up and repurposing or war. And I think you make good points but your assessment is too anchored in WWII. The world, the US economy, and the nature of war are are far different than they were back then.

The US economy is much more technologically sophisticated than it was during WWII. We have a finite capacity for output and if we focus on allocating more of that to manufacturing that is less that is allocated to more sophisticated goods, including hard tech as well as software. There is a finite amount of capital and manpower and if we reallocate towards manuacturing at a large scale that is going to really bite and potentially cut into our edge down farther down the value chain. You should also consider that the sophistication of hard military assets requires a quantity and quality of inputs that is too advanced and diverse to be manuactured entirely in the US. We need global supply chains to produce that stuff. Friend-shoring I can get down with; I think that strikes a good balance between recognizing realities but also appreciating the need to scale up during war.

The power differential between the US and the rest of the world is also not what it was in WWII. The implication of global inequality between countries being reduced is that we would need to rely on allies to win a war, both in terms of producing war assets (e.g. tanks and fighters) and actually coordinating in operations. I don't know if it was one of your posts or somewhere else that I read it, but the statement that we are always preparing for the last major war is evident in your argument.

Similarly, war has become more sophisticated. It's no longer true that it 100% comes down to industrial output. For instance, the US government is one of AWS' largest customers. The edge in war at this point comes down more to things like software than it did in WWII. And that will only become more true as AI matures.

That said, I share your more broad concern. While manuacturing capacity is no longer as important economically or militarily as it was in WWII, the capacity to manuacture hard assets like tanks and fighters still does matter considerably.

I just think the argument you're making needs to adopt more nuance and be updated for the modern context. It also needs to show a greater appreciation for the tradeoffs that would entail. Outside of wartime, reallocating assets from high tech stuff to manuacturing capacity is like selling your mansion to buy a mobile home.

Expand full comment