Introduction

Fields of flowers stretch beyond the horizon as you walk along the warm sunset meadow, gliding your hand along the soft petals. There are no fluorescent office lights beaming on you. There aren’t 20 unread Teams notifications waiting for you. Just villagers gardening, children playing soccer, living in harmony with Mother Nature.

This is approximately the kind of romanticized history sold by anti-establishment writers. Engels, the stepfather of Communism, for example, writes about pre-capitalist, pre-industrial society as if it were some sort of leisurely paradise. And then the big old Capitalist robber barons stepped in, chaining everyone to the factory production line.

The truly miserable part of this pre-industrial leisure was that it was often the byproduct of malnourishment. Angus Deaton describes pre-industrial civilization as having been caught in a nutritional bind: people were too physically weak to work more, but they were physically weak because they could not work to earn more. Historian Rainer Zitelmann estimates that roughly 20% of English and French residents were unable to work more because they were chronically undernourished. In other words, their free time was not liberation as some pop-historians will have you believe; it was caloric underage.

This is a recurring pattern I’ve noticed in modern politics; people hold an incredibly inaccurate and mythicized understanding of the past while simultaneously having a remarkably pessimistic view of the future. “The 1950s were wholesome traditional chungus, medieval villages were cozy warm communities, Neolithic hunter-gatherers were peaceful environmentalists who ate organic food daily, yada yada.”

The truth is, the past was incredibly sick, violent, and despicably impoverished.

We’re Richer Than Ever

We are substantially richer than we were decades ago. No matter which income time series you choose, the broad trends stroke upwards. Real wages, median household income, family income, and so on have all risen over the long run, after adjusting for inflation. Whatever TikTokers and demagogues tell you, the average person today has considerably more purchasing power than the average person did in 1970.

This, of course, doesn’t mean that every single good is affordable, or that every person is thriving, or that every day is guaranteed to be better than the last. Housing is more expensive, healthcare is suboptimal, and there are plenty of bureaucratic nightmares to spill ink over, but overall we are richer than ever.

Early Brutish Societies

Not only are we richer, but we are far safer than ever. Steven Pinker outlines this in his lecture here; essentially, across 20 archaeological estimates, roughly 15% of prehistoric skeletons displayed evidence of violent trauma, e.g., crushed skulls, embedded arrowheads, decapitation, and other features demonstrating pure misery. In contrast, Pinker estimates that only 0.7% of the Western population died from violence. Even after adjusting for the worldwide population, which includes battlefield deaths, genocide, famines, and so on, the figure only reaches 3%. And in the 21st century, the figure hovers at ~0.03%. Quite a long way we’ve come.

As Pinker states, ethnographic evidence spits out similar trends. Early illiberal ‘Anarchic’ societies averaged 524 war deaths per 100K annually. For comparison, 20th-century Germany, which sludged through two world wars and engineered a monstrous genocide, only reached 155 war deaths per 100K. Russia came in at 140, and the United States came in at 2.7.

So hilariously enough, the most brutal, bloodiest, and modern war periods were safer per capita than ordinary life in early societies. And ironically, to steal a conservative phrase, early society was insanely ‘tough on crime’. Eighteenth-century England had 222 offenses that would result in death, including but not limited to poaching and counterfeiting. Step back earlier in time, and you’ll find far more brutish punishments like backbreaking on wheels, burning people at the stake, drowning, blood sports, and stoning.

So in my view, it seems like the greatest privilege today is not just our vast wealth, but the absence of cruelty, the civility of modernity. There was never a random patch update that turned us from moral monsters to civil people, but liberal institutions, courts, commerce, and so on made private revenge less lucrative, and literacy helped expand love beyond just their little tribe. A more universal love (Mozi reference).

If It Ever Shines Again

Somehow, public sentiment seems to move opposite to the evidence. Consumer sentiment charts are currently below levels recorded during COVID and the Great Recession. This is all while unemployment is well below 4.5%, inflation is at a healthy 2-3ish%, and consumer spending on travel is at record highs. Retailers are cycling out inventory so quickly that it’s hard to think why everyone’s so pessimistic.

Pinker’s thesis is that subjective fear influenced by a multitude of effects hardly reflects the objective state of society. People are far more moved by cinematic stories than rows and rows of data. A 60-second clip of Bernie Sanders talking about existential risk and stagnant wages is far more salient, even if it’s grounded in falsehoods. The strange paradox here is that the world can become safer while feeling more dangerous because our ability to spread and document suffering improves faster than suffering can decline. Every little atrocity is recorded and uploaded, every Gazan baby covered in dirt and despair is algorithmically distributed onto every Instagram feed and replayed until it feels like daily life, until it feels like the whole world is about to tip into apocalypse.

And ultimately, the purpose of me writing this isn’t to steel-man the status quo, or to defend the establishment, since I don’t need to defend every bureaucratic mess, or recession, or expensive housing cost to recognize that medieval life was utter dogshit. Progress is not guaranteed, and history doesn’t follow a perfect path to prosperity, but the modern liberal institutions that built society helped escape our brutish past.

The best time to be alive is undoubtedly today, and the next best time is probably tomorrow.

Introduction

Lindsay Clancy slaughtered her three children, and immediately dove for the insanity plea. More specifically, her defense team argues that she was suffering from postpartum psychosis coupled with hallucinations giving her commands. Whether this is the case or not is irrelevant; she is responsible for the children’s deaths. As of August 30th, the jury is still in deliberation.

The hollow epistemology of the internet has handled the case with very little restraint; some pop comments have sourced astrology to explain Clancy’s innocence:

The implication here is that her husband is responsible given his astrology sign. Very convincing stuff. It’s important to remember that the internet is full of horoscope investigations (e.g., Candace Owen’s, Alex Jones, etc.), but beyond the obvious crazies, the whole discourse has touched a new industry: pop-feminism. These are the pseudo-intellectual Instagrammers. The ones who believe that women are uniquely victimized in every walk of life, including the criminal justice system. To the women who feel incredibly disenfranchised by ‘the system’, this seems to be a saving grace, confirming their suspicions, but the evidence doesn’t seem to hold.

Capitalize and Grift

Justine van der Leun is perhaps the clearest example of a pop-‘radical’-feminist taking advantage of the Clancy situation to sell her likely waste-of-time book. But to quickly preface, it’s genuinely insane to me how low the evidentiary standard is in these kinds of spaces. Her thesis is quite simple: women are treated much more harshly during sentencing than men. It’s a bold claim that likely resonates with the unhappy perpetual period (joke) anti-patriarchy crowd. Let’s take a look at the evidence:

Addimando, an abused woman who killed her partner, received 19 years to life, while Jeffrey Epstein sexually abused an unknown but likely high number of minors and received only 18 months?? Fascinating.

There seems to be a massive conflation here, that is, everyone is correct to point out that Addimando’s original sentence of 19+ years was extremely harsh while Epstein’s sentence was obscenely light; however, this is a single comparison that fails to control for relevant variables like the crime, jurisdiction, year, statue, judge, prosecution, evidence, criminal history, plea arrangement, defendant wealth, and quite frankly, literally every relevant variable. So the disparity in sentencing is due to a multitude of factors; thus, we really cannot draw any conclusions on how much, if any, of the disparity is in virtue of one defendant being a woman and one being a man. In other words, comparing two cherry-picked examples tells us virtually nothing about the independent effect of being a woman in criminal sentencing.

I could just as easily tape together a carousel of a woman receiving an absurdly light sentence and compare her with a man who received an absurdly heavy sentence and declare sentencing to be implicitly misandrist. Unfortunately, I’m not interested in mirroring stupidity (joke).

If you take a gander into the comment section, you’ll see the totally unserious horde of pop-feminists nodding in agreement as if any evidence was provided to support such a bold thesis. Raise your evidentiary standard!

Good Evidence

The ideal experiment would be something close to a randomized controlled trial in which judges are presented identical cases while swapping only the defendant’s sex. We’d ideally keep crime, evidence, criminal history, etc static. The fantasy counterfactual.. is not real.

In any case, the next best option seems to be looking at large data sets and narrowing down on cases to control for all of the relevant factors (as mentioned previously), aka, observational research. Ironically, after controlling for available personal and offense characteristics, researchers found that women received sentences that were 29.2% shorter than those of men. Women were 39.6% more likely to receive probation, and among the pool of defendants who were incarcerated (subset), women received sentences 11.3% shorter than those of men.

I’m open to the idea that there’s perhaps an important variable that wasn’t controlled for, which can possibly overturn this result; however, this is infinitely more informative than placing Jeffrey Epstein beside an abused woman’s case. The best available aggregate evidence runs sharply counter to this pop-feminist thesis: among comparable federal defendants, women typically receive lighter sentences.

Funnily enough, I do think that these pop-feminists are sincere in these beliefs, but their deepest confusion is mixing moral outrage, perhaps from personal experience, with causal inference. “Epstein received a disgustingly light sentence; therefore, men are favored in court.” Imagine if we applied this utterly deranged methodology elsewhere; grab an old guy from McDonald’s and one marathon runner who died at age 35 and declare McDonald’s as ‘life-lengthening’. Causal inference is not their passion.

Conclusions and Implications

Lindsay Clancy might’ve been experiencing psychosis when she was slaughtering her children, who knows, but this shouldn’t be used as an opportunity to sanitize the deaths of three innocent children just because she was profoundly sick. Furthermore, this also isn’t the right time to lower the evidentiary standard, as many on the internet would like with their astrology and Instagram carousels. If you want to touch reality, then you need to raise your standard of evidence instead of falling for every emotionally charged social media post.

Today we’re going to discuss his book, Cronyism: The Rise of the Corporatist State. It’s essentially the economic history of the US between 1849 and 1929. He’s also written about Cronyism between 1607 and 1849 in his earlier 2021 book, but we’re here mainly to discuss his newer book, which focuses on Corporatism.

Introduction: Should We Fear The Flock?

2026 is the year of the Flock Camera. This nifty solar-powered roadside surveillance tool is apparently the first step towards 1984: towns are canceling contracts, masked ‘vigilantes’ are pulling up with saws, and activists are acting like Ultron drafted every street lamp. The anti-Flock pitch essentially boils down to “I’m fighting tyranny, not destroying property”. This might have some merit to it, but broadly the justification and vague platitudes surrounding Flock camera debate run ahead of the available evidence.

The most notorious example being a Florida police officer who used Flock’s database roughly 700 times to track his estranged wife. Other officers have purportedly used Flock’s database to do a quick search on ex-girlfriends, relatives, and so on.

This is hardly surprising. Give a trigger-happy industry an on-demand map of everyone’s car and eventually some officer will run a taxpayer-funded audit of his marriage.

On the other end of the debate, though, a new working paper that studies 216 police departments finds that Flock deployment was associated with a ~11% reduction in automobile theft.

So if even part of the estimate here has external validity on a national scale, then Flock seems to be an efficacious anti-theft tool. However, the issue still stands. Parroting either fact is not really doing the necessary utilitarian calculus.

Flocking Hysteria & Framing Effects

The way Flock cameras operate is by tracking and noting down passing vehicles’ make, number, location, and time. Police can then receive alerts if a plate matches an amber alert or any other relevant triggers. In other words, it’s pure surveillance compressing thousands of scattered observations into a working database for nosy cops. Sounds a bit scary.

Lots of people have suggested this would be functionally equivalent to the government sticking a 24/7 tracker on every vehicle, and they are partially right. Flock tracking is not literally like a GPS tracker, since it doesn’t continuously track a person’s location, only once you touch the camera’s view window, and it won’t tell the police where your trip began, where it ended, where you stopped, etc. But critics are correct in that if Flock cameras expand further, the functional difference between this and a government-assigned GPS will be next to none. More clearly, the distinction I’m gesturing to is that surveillance capacity isn’t a binary of 1984 and libertyland; it’s more on a scale between the two.

It seems to me that much of the hysteria around Flock cameras comes from framing effects or loose analogies online. You can test this out for yourself. I’ve tried some less cinematic thought experiments on people who are ambivalent on the Flock issue: imagine hiring thousands of perfect superhuman traffic cops with photographic memory and perfect coordination. They spend the day jotting down plates, times, etc. This is functionally identical to Flock, but it sounds less like 1984 and more like bleeding taxpayer dollars to overpolice traffic.

It seems less relevant to me which analogy produces the strongest scare, but what’s more relevant is how often the Flock database is abused, whether the public safety benefits exceed the cost, and whether it can be reformed rather going balls deep in either direction.

Some Libertarian Objections I’ve Seen

As expected, Libertarians have gone catatonic over this issue. Connor O’Keefe at the Mises Institute, for example, has offered some skepticism but, in my view, transforms the whole discourse into a normatively loaded fever frenzy.

Like many other libertarians, O’Keefe correctly points out the various police abuses I outlined earlier, and also touches on the potential for future government abuses snowballing into a genuine watchman state. The issue is then that he omits the measured benefits of Flock, as well as the empirical infrequency of abuses. He turns the entire policy analysis into an “America’s greatest hits” video where he traces out various abuses, e.g., the PATRIOT Act, the Iraq war, bank bailouts (not even really an abuse lol), and COVID lockdowns. If your argument against traffic cameras hinges on litigating the Iraq war, then I think your band of comparison might’ve stretched to the point of snapping.

The interesting thing is that Libertarians at the Mises Institute like O’Keefe seem to acknowledge that public outrage isn’t really motivated by current misuse, but more by the potential for future misuse. This isn’t exactly an irrational concern since the PATRIOT Act and Snowden’s whistleblowing should update us in favor of the hypothesis that government officials are slightly hiding the scope of their surveillance and that “just trust us bro” is an unsatisfactory standard.

But instead, distrust should raise our evidentiary standard; thus, simply conceiving of a possible dystopian future with Flock doesn’t seem informative about the probability of such an event, and it reduces most of our policy analysis to dreaming of 1984 fanfictions and cutting down any semblance of utilitarian calculus.

The Utilitarian Calculus

There seems to be a lot of uncertainty about the frequency of Flock database abuses, but it seems clear that the frequency is low. Thousands of police agencies use Flock, and incomplete public audit logs show more than 242 million searches from 2021 to the present. Reports have identified over 50 officers improperly using these plate ID systems, with 46 of these involving Flock data. So a lower-bound estimate is 46 divided by 242 million, which comes out to 1.9 E^-7, or, in other words, infinitely small. There may still be abusive officers out there going fully undetected, doing hundreds of abusive searches a month, but that’s why I said this is only a lower-bound estimate.

Ultimately, I’m not at all moved by one-off or highly infrequent examples of abuse, especially when those abuses can be mitigated through restructuring incentives or increasing disincentives (like punishment for abusing Flock data). We certainly lack the evidence to call Flock abuses a national scandal.

Even on the point of “Flock camera errors”, this too seems empirically infrequent or uncertain. Flock claims a 93% correct reading rate. The Institute for Justice claims that there were at least 27 instances of serious errors with 1/3rd of them involving machine errors and the remaining 2/3rds being human error.

More importantly, what do we get in return?

Flock’s own internal estimates claim that the tech helped locate roughly 10,000 missing people in 2025, but the figure seems to be an extrapolation using customer survey data.

Richard Hanania has also made a pretty strong claim in favor of Flock cameras, pointing out that ~120,000 cameras in operation have correlated with a 50% fall in automobile theft. The intuition here is that stealing cars, kidnapping people, and murdering people become exponentially more difficult when your getaway van gets traced every step along the way. I must say, Hanania seems a tad too optimistic and isn’t really pricing in the tail risk of Flock being used to invade privacy and optically demolish dissent. This is also clearly raw speculation rather than causal evidence of Flock’s efficacy; I think the more tame and reasonable claim is that Flock deployment likely reduces automobile theft to some degree.

Richard Hanania points out in a Substack that automobile theft dropped by roughly half as Flock expanded to 120,000 cameras nationwide. The mechanistic intuition is simple: the getaway van for the robbery, the escape car for the murder, and the getaway bike all become far less useful when they leave a trail of breadcrumbs every couple of miles. Although this is mostly a correlation, so you shouldn’t run a victory lap over this, Mr. Hanania. The tamer claim we can make, given our weak evidence, is that Flock contributed to that decline to some extent, but the magnitude is uncertain.

In any case, an immediate moratorium and recall of these cameras is clearly a no-cost task. This, critics, should be careful not to bucket every minor abuse while ignoring the solved murders and missing children.

The more sensible solution seems to be putting restrictions and regulations on these cameras rather than flying into a panicked frenzy and hacking away at them.

Some Sensible Remedies

It’s really great that we’re never in a true dichotomy where we need to choose between unfettered Flock cameras on every street lamp or a survelience free AnCap society. Due to social pressures, Flock has gone from retaining records for 30 days to 7 days. They’ve added required case codes for access, and they have algorithms to flag abnormal searches. It seems reasonable to tack on permanent audit logs which generally don’t take too much data to store and obviously we should be a smidge harsher on people who abuse the Flock database. Perhaps they should get fired and criminally charged instead of just being put on paid leave.

Conclusions and Implications

I still think we should be wary of expansion of government, especially when that expansion involves surveillance.

Shrinking millions of roadside observations into a searchable database seems tyrannical only in terms of privacy violations, and the utilitarian calculus seems to come out positive under a stricter Flock model. It seems healthy to approach government with high levels of skepticism, but it shouldn’t evaporate our evidentiary standard for voicing our support or opposition.

Introduction

It’s easy to understand why famed economist Milton Friedman loved Hong Kong so much. Post-war Hong Kong had low taxes, low public spending, and virtually tariff-free ports. Government regulation was light, and private enterprise was left to its own devices.

As Friedman notes in his essay “The Hong Kong Experiment”, other notable countries like Britain, Israel, etc, were far more statist than Hong Kong, meaning you had a dirty control-treatment split, or natural experiment. Hong Kong, which allowed entrepreneurs to trade, invest, hire, and flourish with minimal intervention, experienced a euphoric surge in economic prosperity. To Friedman, this was undeniable proof that free markets were an integral prerequisite for economic flourishing.

He seems to be mostly correct that Hong Kong was genuinely one of the most open economies in the world, with very little intervention, especially relative to other East Asian countries like Japan and South Korea. Hong Kong never pursued a string of massive state-owned-enterprise conglomerates or state-set production targets; most of Hong Kong’s decision-making was left to the private sector. However, the part where Friedman falls short is in compressing ‘laissez-faire’ to merely low taxes and free trade.

We can trace out some clear counter-examples to show why Friedman’s criteria might not be exhaustive enough to capture how we typically use the word laissez-faire. For example, a government could tax its citizens very little while having full ownership over resources like essentials, major industries, land access, and more. Imagine an economy in which the government owned every plot of dirt, every enterprise and every bank in the country, but had a corporate tax rate of 5%. Is it laissez-faire? I’d be shocked to hear a libertarian endorse such a state of affairs. Thus, it seems highly unintuitive to call such a state of affairs ‘laissez-faire’. But my broader thesis here is that Hong Kong, while a Capitalist haven, was much less laissez-faire than Friedman gave credit for and that natural experimentation here is a myth.

A Brief History of Hong Kong and Its Major Industry

To keep the economic history brief, Hong Kong was originally seized by the British in an attempt to flood China’s marketplace with opium. Hong Kong was a perfect port to connect Britain with an enormous natural harbor. After WWII, Hong Kong saw immigration en masse, i.e., a huge influx of human capital. Shanghai industrialists poured into the streets of Hong Kong, carrying vast knowledge of textiles. This, coupled with the Korean War embargoes, curbed a huge chunk of HK-China trade, forcing Hong Kong into a more export-driven economy, and with it, the birth of a laissez-faire Hong Kong.

By the 1950s, crown-colony Hong Kong was an export powerhouse for textiles, garments, plastic, and electronics. And British Financial Secretary John Cowperthwaite seemed to be interested in economic liberalism for Hong Kong. Cowperthwaite’s successor, Philip Haddon-Cave, held similar sentiments and subscribed to “positive non-interventionism,” which was sort of an ad hoc interventionist philosophy of intervening when markets seemed to fail or when officials believed that there were advantages to intervening (hello Public Choice theory). This seems to cut heavily against Friedman’s “rule-based” laissez-faire philosophy, but it’s mostly a minor issue compared to Hong Kong’s other non-laissez-faire aspects like land ownership.

State Ownership of Land

The biggest problem for the ‘laissez-faire’ narrative is land ownership. Almost all land in Hong Kong, even under crown colony Britian, was owned by the government. The state chose which land to release to the private sector, they chose what the land was to be used for, and the length of the land lease. In other words, they were very loosely centrally planning which houses, factories, shops, and roads could be built where. This hardly seems inconsequential.

There are reasons why you might prefer public ownership, whether it be for egalitarian reasons, externality reasons, or coordination reasons; the point is:

1. This makes it hard to argue that Hong Kong was incredibly laissez-faire

And

2. It cuts against the supposed “natural experiment” of Hong Kong

Public Housing, Manufacturing & Textile Quotas

Public housing is another aspect that doesn’t exactly fit nicely into the laissez-faire Friedmanite narrative. After the 1953 Shek Kip Mei fire, tens of thousands were homeless. This, coupled with the immigration influx, led the government to build swathes of low-cost resettlement homes. By 1965, 1.8 million people (roughly half of the 3.6 million population) were living in government-granted housing units.

This also intertwines with the manufacturing and export industry (credit to my friend here for bringing up this uniquely interesting point). Housing is surprisingly more than just a welfare program detached from the rest of the economy; it’s also a relief valve for labour-intensive export industries. Lower housing and transportation costs mean workers require less money to maintain their standard of living, thus serving as an indirect subsidy to manufacturing. They do all this without spending a dime. This mostly seems like a drop in the bucket since under an economically liberal setup, you’d likely see an enormous rise in the housing supply in response to a demand shock. So the government really was responding to the spike in demand in a similar way to how market forces would have responded.

And even without tariffs, Hong Kong utilized export quotas for textiles, though I would partially blame this on other Western countries being protectionist. Suppose the United States caps Hong Kong’s textile imports to 500 garments a month; an efficient means of rationing this is by allocating export rights based on a firm’s previous export performance, which predictably favors incumbents. So I’m a bit hard-pressed to chalk this up to Hong Kong being interventionist, rather than simply responding to intervention imposed by others.

Banking Licensure

Aside from land ownership, the only other semi-notable component that cuts against the laissez-faire narrative is banking licensure. After the banking sector experienced some instability in the 1960s, the government imposed a full moratorium on new licenses, artificially creating barriers to entry. Similar to the United States, they enacted various restrictions on interest rates and reserve requirements.

Catherine Schneck notes that this was highly anti-competitive, and license restrictions were essentially an inefficient version of prudential oversight.

This doesn’t seem super consequential to me. Considering Friedman was a staunch monetarist and supporter of the Federal Reserve, I’m sure he knew of this, but simply didn’t feel particularly moved by a few instances of messy interventions in the grand scheme of things. It seems perfectly fine to me to say that a central bank’s existence is compatible with a laissez-faire economy, but I know some Austrians who might be particularly moved by this.

Conclusions and Implications

It seems clear to me that Hong Kong, while overall economically laissez-faire in many respects, was less laissez-faire than Friedman ever gave it credit, but it’s also overstated by libertarians trying to litigate Hong Kong’s legacy.

More importantly, it’s incredibly difficult to use Hong Kong as a natural experiment for proof of laissez-faire being optimal relative to a more corporatist system like South Korea. I think perhaps it is weak evidence of laissez-faire outperforming more intervention-heavy systems, but that requires using more analytical methods than Friedman did.

For example, a synthetic control paper on Hong Kong finds that limited government growth had no visible effect on economic growth. I’m sure there’s lots to scrutinize here, but the point is that it’s not super clear what worked with Hong Kong and to what magnitude.

It’s almost a hilarious irony that the best example of laissez-faire was one borne out of Western subjugation of a Chinese territory. So to be clear, I agree that Hong Kong is overall quite laissez-faire, and a few drops of intervention don’t mean you can proclaim that Hong Kong is a corrupted corporatist Mordor.

Introduction

In 2013, Scott Alexander drew attention to a fascinating survey phenomenon: the “Lizardman Constant”. A survey found that roughly 4% of Americans “believed” that shape-shifting reptiles secretly controlled the world. Taken at face value, this would mean that millions of Americans believe Congress is secretly composed of humanoid lizards rocking a suit and tie.

The internet is filled with morons. I think it’s possible that a few people believe something as absurd as this, but for the vast majority of survey respondents, I refuse to believe they sincerely believe this.

The more plausible explanation for this is that among those 4%, people were either joking, confused, speedrunning the survey, or misread the question.

In any case, the point that Scott Alexander is alluding to is that every survey has a noise floor, i.e., some percentage of respondents answering questions in a way the researcher never intended.

And I love how “Lizardman Constant” sounds; it’s a very catchy label, but it’s misleading. When people hear “constant”, the implication is that there’s a noise floor they can just cleanly shave off and magically transform contaminated results into scientific standards. The assumption here is that contaminated respondents behave in a stable, predictable way, such that we can simply shave off a portion of the results. This obviously is not the case.

For example, imagine a troll who answers “yes” to every absurd question because he thinks it’s funny.

“Are you an indigenous transgender guppy?” Yes!

“Are you a telepathic Peruvian hamster?” Strongly agree!

Despite a person’s behavior towards strange questions, they might still answer ordinary political questions with sincerity. Another respondent who’s sleep-deprived and half-running on a can of Monster might just zip through the survey questions to finish as quickly as possible. A sycophant might do the opposite and just agree with whatever they think the researcher would perceive in the best light. Someone who’s lazy might retreat to the middle options. The point is, all of these respondents can be labeled as noise, but they don’t produce the same noise, which makes it very hard to split from the useful data.

Your Results Are Trash! Cope!

For decades, survey scholars have been pulling their hair out in an attempt to split substantive beliefs from the jungle of confusion, fatigue, and button-mashing. As Jon Krosnick argues, answering survey questions requires a fair amount of effort. Respondents need to understand the question, form a judgement, and select one of the available options. They have every incentive to take shortcuts (‘satisficing’), even if they’re just picking something that looks “good enough”. This is doubly true when surveys are long, repetitive, and boring, with non-existent stakes.

Sometimes ‘satisficing’ is not as obvious as some guy answering A for every single question. As I mentioned earlier, respondents have every incentive to take mental shortcuts. Instead of carefully thinking through a question, a respondent might only glance at the first two options and choose from there. They might give themselves a maximum of 5 seconds to answer the question. Or they might give up halfway through.

You can potentially screen for data contamination by implementing specific protocols (Meade & Craig, 2012), but they’re not exactly perfect. Someone who selects the same option thirty times in a row might just happen to have this combination of beliefs, but more likely, the respondent has just realized that spamming the second bubble is the fast lane to the completion screen. Other protocols can identify patterns, e.g., a respondent who answers: ABCD, ABCD, ABCD, ABCD, is probably not answering that by chance.

Researchers can also tailor their survey to test for consistency. For example, if a survey question asks, “Are you afraid of the dark?” and the respondent answers “yes,” and then you, in a later question, ask, “Does the dark scare you?” and they answer “no”, then you’ve spotted an obvious contradiction, which is good evidence that the respondent is not actually answering sincerely.

The issue is still that no researcher heuristic can perfectly identify low-quality respondents, so it becomes a game of “can you justify your research design and data omissions” without looking like you just retrofitted your preferred ideology onto the survey data.

The Lizardman Ruins Everything

This would mostly be a non-issue if low-effort respondents were tiny blips in research data; unfortunately, it’s quite the opposite. As Huang, Liu & Bowling (2015) note, high levels of brainless respondents can create relationships that aren’t there. Imagine a personality survey with 20 questions; a sycophant clicks “strongly agree” to every single question. The thing you’d see from this, if you didn’t know any better, is that all these traits are correlated with one another. So the guy spamming A-B-A-B-A-B with his forehead while stimming off a Red Bull doesn’t notify the spreadsheet that you’re getting spoonfed contaminated data.

The scary lizardman “constant” is especially dangerous when you consider how tiny percentages can be extrapolated onto the national base. Imagine if 48% of respondents show support for a policy, an estimated 2% contamination problem is mostly a non-issue for making claims like “at least a quarter of Americans support policy X”. But sensational headlines are salient. A small percentage of people supporting something translates into millions nationwide. Three percent of people support segregation? That’s ten million people!

I’m not even joking; this happens so frequently online you’d think we would have publication rights by now. A 2014 paper, for example, used Coop Election Study data to estimate that 6.4% of noncitizens voted in the 2008 presidential election. The authors then multiplied that figure across 19.4 million non-citizen adults to “discover” that over a million illegal votes had been cast, which arguably shifted election outcomes. A 2015 reanalysis found that there was a citizenship-classification error of merely 0.1%, which gave rise to this insane extrapolation.

Messy Messy Messy Biases

Even with a perfectly sincere pool of respondents, it’s not so clear that your survey data will be contamination-free. A person genuinely answering in good faith can still be influenced by a modicum of subconscious biases.

I wanted to do some admittedly messy testing on position effects. To do so, I ran one on Substack and one on my friend’s Twitter to minimize audience overlap. In both polls, I made every option identical, and in both polls the third dot won. This, of course, isn’t some high sample size randomly selected survey, and you could probably prosecute me for polling malpractice, but the only point I’m driving here is that respondents in this convenient sample did not distribute themselves evenly.

The boring null hypothesis here is that absent any peculiar aesthetic preferences, each option should receive a 25% voting share. Using a chi-square (goodness-of-fit) test, we can see the Twitter poll had a chi-square statistic of 8.86, while the Substack poll had a chi-square statistic of 6.10. Pooling the results together, we get a distribution that is highly unlikely under an equal-choice model (X^2 (3, N=124) = 12.58, p = 0.006). In other words, the third dot received 17 more votes than the null hypothesis predicts.

Ironically, though, this outcome is pretty consistent with the literature on positional effects for polling. Nicholas Christenfeld examined how people make choices under identical options and found that people mostly favored middle options over the top or bottom choices. In one experiment, he found that having identical products on a shelf resulted in the middle items being the majority of selections. This obviously doesn’t mean option 3 is universally favored, but it’s to say that there are peculiar biases that make clean survey adjustments nearly impossible.

More broadly, even when answering very sincerely, ordering, framing, and other factors can raise the probability of a respondent choosing an option they wouldn’t have chosen under completely neutral conditions. Imagine someone sincerely believes option 2 is correct; the likelihood of them choosing options 1, 3, or 4 is low but non-zero. Maybe option 1 has a “first-comer” advantage since it’s at the top of the list. Maybe option 3 just looks appealing due to its position. Maybe option four looks aesthetically pleasing.

And of course, you can’t solve for this by just collecting a fatter sample, since you’d just be amplifying the amount of noise in proportion to your sample. Directional biases are truly awful. But all is not lost; you can still perhaps slay the lizardman!

Proposed Solutions in the Literature

There seem to be a few interesting proposals for hedging against the “Lizardman constant”.

The most convincing to me is just simple randomization. If researchers shuffle the order of options across respondents, then on a systematic level, it seems like any kind of positional or cognitive effects will be distributed relatively evenly across options. The same would go for salience effects; it seems obvious that researchers shouldn’t make one option bold or one option bright red, so if you treat them all the same, you should be able to minimize salience effects.

As mentioned previously, you can combine quality indicators, whether it be recognizing straight-line responses, checking for repetition, contradictions, and so on.

On top of this all, replication, which is standard research practice in economics, medicine, etc, is a fine way to validate results and test for sensitivity. Perhaps if you tweak the survey question just a smidge, you get wildly different responses.

None of these guarantee the truth of trying to gauge public sentiments, but at the very least, good research coupled with minimization strategies should provide a strong abductive model as to what public opinion leans towards.

I also find it a bit interesting how half the survey ecosystem leans on public support. People are quick to cite polls when it vindicates their views, but less so when surveys find something boring. So that’s maybe another potential conflict of interest to explore more in the future.

Conclusions & Implications

So to sum up, the Lizardman Constant isn’t really much of a constant; it’s more so a swath of cognitive failures that are near impossible to split off from the sincere survey data. There’s no easy correction to all of this. You can do tedious things like randomize, model misclassification, etc, but you end up in the same boat where you’re trying to justify your modeling choices or omissions. So it seems reasonable to approach surveys with a high level of skepticism. In practice, this means you won’t ever spit out one clean authoritative number, but perhaps you’ll cluster around something useful. If something polls at 40%, the Lizardman is probably a blip of noise, but at 4%, he might just be the entire result.

Introduction

“The optimal amount of crime is greater than zero”. On the surface, this seems a bit, well, insane. Why would anyone want any level of crime? Surely, a world devoid of murder, theft, fraud, & violence is better than a world with this? At least, in some possible world, in some perfect utopia, the optimal level of crime should obviously be zero.

I think the deep confusion starts with how we view crime. Crime is a legal category, not a yardstick indexed to social utility. If the government went full nutjob anarchy and abolished legal enforcement & the criminal code, then the official crime rate would drop to zero by definition. We wouldn’t have eradicated murder, or theft, or assault; we just wouldn’t acknowledge them as crimes. So while the statistic of zero crime might look prima facie good, society would undoubtedly be far worse off.

Secondly, when we claim some level of crime is optimal, the word optimal seems to be indexed to some conception of aggregate social utility (welfare). I won’t dive too deeply into conceptions and arguments for interpersonal utility comparisons and aggregation problems, but feel free to check this out if that interests you. In any case, everyone would agree that minimizing crime is good, ceteris paribus. However, we face trade-offs.

Untangling the Semantics of “Optimal Policy”

When an economist says policy A is optimal, they do not mean that it is the best metaphysically possible arrangement. They mean something far simpler: that is, this policy is the best option among the set of alternatives. Suppose a financial analyst describes a stock portfolio, B, as optimal. They do not mean that it delivers an infinite return with a portfolio volatility (standard deviation) of zero. What they mean is that, given the asset combinations available, portfolio B offers the best available risk-return combination. The same intuition should be applied to the semantics of “the optimal level of crime”.

“The optimal level of crime is greater than zero” can be interpreted as “given the available policy options, the policy that produces the highest level of social welfare has a level of crime that is non-zero. That crime is deliberately unprevented.”

The implication here is that the government doesn’t select a crime target; there’s no greybeard deciding that Baltimore needs between 200 and 300 murders this year. The implication is that a positive amount of crime is a residual amount that would be too costly to prevent since the social harm of the crime is less than the cost of preventing it. So actually flattening the crime level out to zero would likely take an extraordinary amount of tax dollars for universal surveillance, preventive detention, 12-digit police budgets, warrant-less searches, and severe capital punishment. You can draw a parallel here with healthcare debates; we could probably reduce mortality and increase healthcare access by pouring cash into hospitals, medical equipment, etc, but at some point the marginal gain from each dollar spent would be so negligible that it undoubtedly would be better spent in another sector, i.e., housing, research, infrastructure, etc. So it seems once you introduce a few extra qualifiers, it becomes abundantly clear that the crime level being non-zero is optimal.

Becker’s 1968 Puzzle

Gary Becker famously posited this question in his 1968 paper Crime and Punishment: An Economic Approach. In it, Becker questions “how many crimes should be allowed and how many criminals should be left unpunished”. The idea is simple: we want to identify the combination of enforcement and punishment that minimizes aggregate social loss.

To simplify Becker’s model, suppose the function N(e) represents the number of offenses at enforcement level e, where more enforcement means less crime. So N(1) > N(2).

Now imagine we have the function H(N(e)) representing the harm associated with the number of offenses at enforcement level e. And let C(e) represent the social cost of enforcement; think taxation, court costs, privacy violations, wrongful convictions, police brutality, and so on.

Thus, we have the social-loss function:

L(e) = H(N(e)) + C(e)

Where the policymaker can control the level of enforcement, e. The goal is to choose a level of enforcement that minimizes L(e), which is found by setting the derivative (the rate of change) to zero, corresponding to the turning point or global minimum. The function is represented by:

dL/de = H’(N(e))*N’(e) + C’(e) = 0

Or

C’(e) = -H’(N(e))*N’(e)

Without all the mathematical sludge, the equation can be interpreted as “as long as spending another dollar on enforcement (e) prevents more than a dollar’s worth of offenses (social harm), then spending is good. Once the marginal cost of enforcement exceeds the marginal cost of crime, further enforcement makes society worse off. Therefore, the optimal point is where the marginal cost of enforcement = the marginal harm of the crime.

The practical implication here is that one additional dollar on enforcement (police) might prevent several crimes through sheer deterrence. It might increase the chance of catching more crimes. It also might just have no effect.

Imagine if the first million dollars in additional prevention stops 1,000 burglaries, the second million prevents 100, and the third million prevents 10. If each prevented burglary is associated with a $20,000 social benefit, then the first million in enforcement would net $19 million in benefits, the second would give you $1 million in social benefits, and the third million would give you a net social loss of $800K So a policymaker who decides not to burn that third million on extra police might realize that preventing 20 more burglaries isn’t worth $1,000,000, meaning the optimal amount of crime is non-zero, as unintuitive as that may sound.

Conclusion

I was a bit taken aback hearing that “the optimal level of crime is non-zero”, but once you dissect the claim a bit, it seems abundantly obvious. Reducing crime requires resources, and at some point, the next dollar spent on enforcement is equivalent to flushing cash down the toilet. Thus, under realistic assumptions of costly enforcement and diminishing returns, the optimal level of crime is indeed above zero.

Few men of the 20th century truly earned that “polymath” label as convincingly as Ludwig von Mises. His expertise stretched from economics to politics, to philosophy, to history, giving his work unmatched breadth and depth. But eventually, great thinkers have to answer to time, to contemporary evidence they never lived to see.

In Human Action, Ludwig von Mises provides an account of Rome’s civilizational collapse from an Austrian perspective. Mises argues that currency debasement & price controls paralyzed food merchants and crushed the division of labor, turning Rome into a sort of autarkic (non-cooperative) production process with little urban commerce. Mises seems to be trying to explain the crumbling of Rome more broadly, rather than just the capital, but this account also supposedly explains the internal struggles that made invasions possible.

Mises writes:

“What brought about the decline of the empire and the decay of its civilization was the disintegration of this economic interconnectedness, not the barbarian invasions.” [...] “(price controls & debasement) completely paralyzed both the production and the marketing of the vital foodstuffs” (Human Action, page 762)

The two distinct parts of this are:

1. Currency debasement hollowed out the Western Empire over two centuries

2. The 301 price edict did enough damage to outweigh competing explanations

These are very strong claims that require more than theoretical mechanisms. The analysis is certainly elegant, intuitive, and interesting, but it is far too reductive and convenient. And to be clear, I oppose price controls. I have no doubt that price ceilings discouraged production to some degree, and debasement certainly made entrepreneurial planning difficult, but it seems like an incomplete view of history. In Mises’ analysis, several notable factors go unmentioned or are subordinated to economic discussion, meaning the politics of the invasion are left undiscussed.

For starters, the actual Diocletian price edict occurred in 301 AD, whereas the Gothic (barbarian) sack of Rome occurred over a century later in 410 AD. If price controls were so damaging, it seems strange that it would have taken over a century to fester into a disaster. Furthermore, it’s not so clear to me that the price control was all that enforceable, meaning it was a price control on paper only. As for debasement, over the span of 2 and a half centuries, silver content dropped from 90% under Augustus to 5% by the 3rd century. It seems also unexpected here that such enormous debasement would produce no immediate economic rupture, but a catastrophe only in the 4th century AD.

There’s also a messy natural experiment we use as weak abductive evidence between the East & West. The Eastern Empire (Constantinople) inherited all the bureaucratic and statist baggage that the West did, yet it survived ~a millennium past the sack of Rome. This would certainly be highly unexpected under the Misesean hypothesis, where statist baggage is an enormous driver of imperial collapse.

Not to sound too ‘empirically-minded’, but I think the obvious issue in all this is a lack of an identification strategy. A century-long gap between the price edict and the sack of Rome doesn’t disprove Mises’s conclusion, but it seems to be evidence that lowers the probability of his view being true. And remember, the burden is on him (or other Austrians) to show that price controls were the reason for the persistent decline in commerce and the weakened military capacity.

Keep in mind, Mises’ empirical scaffolding leans on Michael Rostovtzeff’s book, “The Social and Economic History of the Roman Empire” (see page 762 footnote), published in 1926. Because Mises wrote Human Action in the 1940s, his pool of resources is far less comprehensive than the work we have in 2026; thus, given how contemporary scholarship has focused on archaeology, regional comparisons, and ceramics, the evidentiary standard seems to have grown far past what Mises could have even possibly provided in his nearly century-old book.

The irony here is that Mises is correct that long stretches of food networks made Rome vulnerable, but the economic collapse in 410 AD was not from merchants refusing to sell grain under Diocletian’s price ceiling; it was the army that controlled rivers, harbors, and other trade routes. So skipping over the military land control while elevating price controls as a factor seems mistaken.

The more plausible story seems to be that prior to Diocletian, there was massive political instability. Emperors would get shuffled in and out in mere months. The implication here is that virtually every policy gets slashed before it bears fruit. Thus, Diocletian, who served for decades (284-305), stabilized government volatility through the Tetrarchy, designing a four-emperor autocracy. As you can imagine, as is the case with most autocratic regimes, dissent is inevitable.

So in essence, Diocletian’s authoritarian ‘Tetrarchy’ temporarily stabilized political regime continuity, only to violently collapse into self-sabotage. After a wave of mass paranoia swept the elites, Honorius (the Western emperor) ordered Stilicho’s (a general) execution. Roman soldiers subsequently slaughtered the wives & children of barbarian federates living inside Italian cities, which caused the surviving relatives to defect to the Barbarian army outside the city walls. Not the smartest move to say the least.

It certainly seems incomprehensible that a century of price controls would have set the stage for the invasion of Rome; after all, the defensive infrastructure was incredible. The walls were mixed with volcanic ash and lime, standing at 8 meters high and 3.5 meters thick, reaching ~12 meters high by 403 AD. The Gothic barbarians could surround the city, cut off food supply lines, or wait for an unlucky famine, but actually vaulting past thick slabs of limestone is an entirely different matter.

The story that I most believe is Procopius’ account, who claimed that 300 Gothic youths entered as enslaved ‘gifts’, only to kill the guards and open the Salarian Gate. Sounds like a plausible Trojan horse story. Another plausible explanation that I’ve suggested is that Goths already living inside the city secretly worked with outside forces to open the gate. In any case, the historical consensus seems to be a mix of siege and internal fracture, which cuts against Mises’ anti-statist account.

While I think Mises is correct in observing the pain caused by price controls, he falls extremely short by blindly assigning causes without considering other non-economic factors such as a lack of Gothic assimilation and political fault lines. Even if he’s correct that debasement strained commerce, the invasion in 410 seems to be explained better by failed Gothic assimilation, military fractures, and a government that slaughtered the militant coalitions that it needed to ward off invaders. We can imagine a counterfactual in which prices were liberalized, leading to more prosperity, but this seems unlikely to have prevented inter-coalition slaughters and mass defections to the barbarians.

This is a frequent problem I’ve stumbled on in heterodox economics. Because these schools badly want a paradigm shift, they constantly market themselves as deeper alternatives to the empty world of Neoclassical economics. So it’s no surprise that heterodox thought leaders feel so obligated to not only explain more, but to explain everything with their theory. Where Austrians find price controls behind every empire’s collapse, Marxists find ‘internal contradictions’. So in short, Mises may be right in pointing out that Diocletian’s price edict made the empire poorer. In fact, the economic story might just be causally upstream of the political explanation, but economic intervention didn’t cut the city’s militia or create political dissent; thus, Mises is smuggling in a strong causal thesis with a very thin layer of outdated evidence.

Introduction

Everyone knows New York City is the beating heart of American Finance. Walk down Wall Street, and you’ll see Goldman Sachs, JP Morgan, Blackstone, and so on. Every finance bro is there with their silky North Face vest. But why New York City?

The obvious first thing to check would be pro-business policy. Maybe, just maybe, New York keeps its corporate tax low, maybe it has a low regulatory burden, and maybe it’s strong on private property. Nope. On every relevant metric, New York City is a glowing blue state with high taxes and a literal socialist as Mayor. Major corporate conglomerates are getting hit with a 9% city corporate tax on top of the state and federal taxes. Imagine if you were an alien examining the optimal spot for a big fat Goldman Sachs, looking solely at public policy logistics around the country. Seems unlikely you’d pinpoint New York City as such a bustling finance hub.

Yet New York City is the global finance hub. I suppose the deeper question is ‘how do financial centers emerge?’ Certainly not through low taxes alone. If that were true, all these firms would be shuffling over to Alaska in a moment’s notice. My thesis is that New York City became the dominant finance hub because of its historical strong points, which gave rise to rapidly developing commercial areas, infrastructure, network effects & massive knowledge accumulation. In modern times, New York City’s position in finance is anchored by the neighborhood client base.

Commerce & History

The biggest advantage of New York was its natural harbor. Long before any skyscrapers were erected, Manhattan was nested beside ports protected from waves but with direct access to the Atlantic. The Hudson River connected the city’s interior, and the Erie Canal stretched to the Great Lakes and down to New York.

As Economist Edward Glaeser notes, technological progress nudged the city away from commercial shipping. Harbors made expensive ships lucrative, and with ships, brought merchants. With merchants bringing insurers, storage providers, lawyers, and the whole finance industry to fund the operation. Capital-intensive activity meant merchants needed someone to extend them credit, someone to plant that seed capital to fund their project. Ship owners needed investors to fund long voyages and insurers to hedge the insurmountable risk of a storm crushing all their loot. So you can kind of already see the picture and see why all these industries required one thing: deep financing.

Once enough people were squished into the same metropolitan area, financial institutions bloomed. As Glaesner notes, the 1792 Buttonwood Agreement built out infrastructure that lowered financial market transaction costs. Instead of praying that the counterparty would hold up their end of the deal, financial markets became more organized, leading ot higher trading volume and more trust. And per the NYSE’s history, by 1817, trading became high-volume enough to expand financial infrastructure.

Eventually, New York displaced Philadelphia as America’s dominant financial center. Glaesner writes that the main reason for this was that New York had far more access to British capital. New York was also a much larger land mass, meaning more companies could be squeezed in. More importantly, financial markets now reward people for allocating capital efficiently through abnormal returns on their investments.

Follow The Client

I’ve worked in a lot of finance client-facing roles, and I must say, half the battle is impressing the client implicitly with your lavish office that just smells like professionalism. We had a nice water fountain in the waiting room. But more importantly, we wanted people to feel comfortable forking over half their life savings and sensitive information. I think we can a priori know that a client is far more willing to meet with you when doing so doesn’t require them to take a $600 plane ticket to North Dakota, so the other issue is being in close proximity to clients.

Outside of just pure asset management, this is doubly true for investment banking/underwriting. No one is going to trust you after a couple of exchanges through Outlook or Microsoft Teams. A corporate executive choosing an investment bank will want to meet the people handling the underwriting. Historically, the clientele lived in NYC. I’m talking about the shipping companies, manufacturers, global exporters, and so on. So this element makes NYC’s financial dominance quite expected.

As Pazitka & Urban note, after looking at 1.7 million investment bank interactions across 540 metropolitan areas between 1993 and 2016, financial centers with deeper connections in other cities grew exponentially faster. The implication here is that you’re competing mostly on domestic clients, but also on international clients.

This partially explains why Goldman and JP Morgan are clustered in the same district. In the econ101 model of competition, two competing firms will need to fiercely fight over a small pool of customers, e.g., a McDonald’s and a Burger King planted 100 feet from each other, essentially trying to survive & kill each other. JPM and Goldman, on the other hand, need each other to form syndicates. Goldman might need five other investment banks to help it underwrite and distribute a huge bond offer, and they’ll reap none of the benefits.

In any case, this seems to have far more explanatory power than any overly simplistic explanations like “well, investment banks cluster in NYC because the clients like the vibes.” New York might not even be the optimal place to set up shop, but all NYC really needs to do is stay valuable enough that the pain of leaving is greater than the pain of staying.

A Spoiler-Free Verdict

Spider-Man: Brand New Day was absolute kino: a vibrant color palette, a modicum of POIs, an energetic cast, rapid character development, philosophical flourishes, and a handful of intriguing mysteries. As shown in the trailer, one of the movie’s unique features is peering into the mask as Peter mounts his highest-stakes moments. But this isn’t to say the Movie was perfect; in fact, there are several criticisms I have of this film. I give it an 8/10.

The Trailer Ate the Mystery (SPOILERS AHEAD)

Before walking into the film, I already knew the story would feature OG Hulk, The Punisher, Damage Control, etc. The unfortunate part is that OG Hulk, who hasn’t made an appearance since Thor: Ragnarok, would have made for an insane reveal had the marketing team not treated every surprise as promotional ammo. Instead, I was left with a half-filled road map where I knew exactly what was to come. By the time Jean Grey mentioned ‘visiting Empire State University,’ it was clear that she was going to mind-control him into becoming the Hulk.

So imagine if you came into Avengers: Endgame already knowing that all the vanished heroes would be revived at the end to fight Thanos’ army. What a disgustingly awful spoiler. Though I understand why the marketing team would do such a cruel thing. OG Hulk dragged along an enormous amount of hype with it.

So the predictability is really my main criticism of the film, which is really no fault of the film; rather, it was the fault of the bastard marketing team. And to reiterate, the movie was fantastic.

Identity Crisis

The entire story begins in the aftermath of No Way Home, where every trace of Peter’s existence has been wiped from memory. To be completely honest, Peter’s new secluded life is pitiful at best, and suicide inducing at worst. He has no family or friends anymore. His response to losing his day-to-day identity is filling in the gap with full-time heroism. It sort of parallels how workaholics operate. Their personal life withers away, so they substitute it with 24/7 productivity.

The first-person swinging sequences where we see under the mask capture this theme quite well. Instead of seeing a red suit bouncing across the city skyline, we see things from Peter’s POV. Every height becomes way more real and intimidating, every headache more visceral, and every tear a little more harsh. I have no doubt this was the goal of the directors, considering the entire premise of his identity crisis is becoming less human and more arachnid, which causes his enhanced senses to amplify further.

Cruelty

To make matters worse, the directors buried Peter under mountains of emotional damage. In the beginning of the film, they kind of pump-fake you into thinking there’s some sort of hope for Peter & MJ, but it quickly gets crushed in both instances it is hinted at.

The first instance was when Peter stumbled his way towards MJ at a house party. They exchange a short, cordial chat, and then he sees her kiss her new boyfriend. The sliver of hope is stomped on just moments after it appears, which triggers a silent emotional crisis for him as he swings around, devastated in nothing more than a hoodie. It’s at this point the movie hints at a suicide attempt, or at the very least, flirts with the idea of Peter just shattering onto the ground.

This happens a second time too, except this time it’s far more viscous. It’s quite difficult to think of something more harmful to Peter than this. Jean Grey, a secondary antagonist and a mind controller, uses MJ to get close to Peter. She lets Peter believe that MJ actually remembers him, and for a few seconds, it feels like he got his old life back. Jean lets him touch MJ, his happiest fantasy, and then quickly tears it away while laughing and throwing MJ off a 10-story building. Jean doesn’t even try to hurt Peter physically, just everyone he cares about, which is probably far worse.

MJ is Not MJ

The first philosophical flourish I noticed was the scene in Punisher’s ‘man cave’. After MJ reads Peter’s letter, she rejects his love. She understands what Peter saw in that MJ, but refuses to inherit any of that MJ.

This seems quite like the psychological-continuity account of personhood, in which MJ still has the same body and cells, but the memories, experiences, and connections have evaporated. The way she talks about that MJ or that girl, who is 1:1 physically identical (genetically) to her, is incredibly distant. From her perspective, the MJ Peter wrote the letter to is another person.

I think this is a pretty good storyline. MJ cannot love him because she doesn’t know him. This MJ is a different MJ than the one we saw in the previous 3 installments. So if Peter & MJ are to reunite, it’ll be through a fresh connection, since the connection with the ‘other MJ’ is permanently closed.

Spider-Man is Not a Utilitarian

I hate the no-kill rule for superheroes. Metzger, the director at Damage Control, experiments, kidnaps, and kills mutants. The consequentialist case for allowing Jean to kill him seems abundantly clear: one dead director means way fewer victims down the line. Because of Peter’s moral restraint, Metzger literally escapes.

And another thing. There seems to never be a fatality in any of these enormous battles. Somehow you can shatter a square kilometer of NYC, barrel an armored truck across 5 boroughs, and smash a guy into a wall with zero net fatalities. I just feel like keeping the death toll fixed at zero kills any kind of high-stakes immersion.

But despite all these shortfalls, I enjoyed Brand New Day a lot. Exploring Peter’s identity crisis is a unique storyline that gives every element of his costume a reference to something deeper. And if there’s something I can appreciate, it’s a clean ending that doesn’t open more questions than it answers.

How do the rich avoid taxes? The most prevalent explanation is “Buy, borrow, die”. The clean, intuitive mechanism is as follows: buy appreciating assets -> take out loans using your assets as collateral to fund your consumption -> Die before selling your stock, effectively dodging the capital gains tax -> your heirs then escape taxation through a ‘stepped-up basis’.

Somehow, “buy, borrow, die” has managed to reach the pinnacle of political discussions despite its empirical infrequency. The implication is that wealthy people have virtually no taxable income and receive a perpetual stream of tax-free funds to prop up their lavish lifestyles. This all sounds quite plausible. More importantly, it’s something that does happen legally, just not frequently.

However, saying it happens or saying it’s possible is much different from the claim that wealthy Americans frequently use this to finance their consumption. To dig a bit deeper, we’d want to know how much they borrow in aggregate, rather than just blindly looking at nominal figures or instances of taking out loans. Does the debt accumulate and cause financial distress? In short, the popular narrative of “buy, borrow, die” doesn’t hold up.

The Myth of “Buy, Borrow, Die”

Zachary Liscow and Edward Fox, in their paper on “The Role of Unrealized Gains and Borrowing in the Taxation of the Rich,” cover pretty much every relevant point on the topic.

Because the Fed’s surveys on consumer finances oversample wealthy households, detailed data on the asset classes of wealthy individuals are incredibly transparent and accessible. They compare economic income, i.e., the total amount you gain financially over a year, with taxable income, i.e., the amount gained that is actually subject to taxation. So, for example, if you had a taxable income of $50 and an unrealized asset appreciation of $50, your total economic income would be $100, but your taxable income base would be $50, meaning the government captures 50% of your economic income.

The question everyone is dying to know is: how much does the regular old income tax miss from the top 1%? For the wealthiest 1% (2004-2022), or a net worth of more than $14 million, taxes captured ~60% of their economic income. For the top 0.1%, taxes captured ~50% of their economic income.

So it seems abundantly clear that yes, a sizeable portion of economic gains by the ultra-wealthy escapes taxation. However, the question in dispute is whether this is done through “buy, borrow, die” or something else entirely. Spoiler, the answer is something else entirely! If the rich & ultra-rich really are converting unrealized gains into spending money, then this should be reflected in their balance sheets.

The truth is, though, this is nowhere to be found. Net new borrowing from the top 1% hovered around 2% of total economic income in 2022, while increases in unrealized gains floated at a whopping 41%. And for the top 0.1% (’ultra-wealthy’), net new borrowing was ~1% of economic income while new unrealized gains were 50%. So why all the confusion?

Flashy Headlines are Scary

This section is less so empirical and more of just my speculation as to why “buy, borrow, die” has gained so much notoriety despite its sparse use. The top 1% (excluding the Forbes 400) carried ~$1.02 trillion of debt in 2022 alone. This enormous figure is then plastered in every Yahoo Finance article, every Reddit forum, and every debate space. The scary headline spreads like a pathogen, and the ones spreading it seemingly never realize that nominal figures aren’t good evidence.

In any case, the number to compare that nominal $1 trillion figure to is the $23 trillion in unrealized gains, which makes the household debt only 4% of accumulated unrealized gains. That 4% of debt-holding isn’t being deployed as a way to fund consumption and avoid taxes. As Liscow and Fox find, ~66% of the top 1%’s debt is nestled in residential real estate, ~12% is in loans to leverage securities, and the rest is bucketed in various lines of credit and ordinary consumer borrowing. The authors go a step further and do the thankless task of simulating a scenario in which consumption is funded through asset sales instead of borrowing. They discover that this would only raise the top 1%’s income tax rate by a meager 0.2%. A great nothing burger.

There’s No Mountain of Debt

Another glaring inconsistency in the “buy, borrow, die” hypothesis is the debt empirics. If the wealthy & ultra-wealthy actually perpetually borrowed cash against their assets to fund their lifestyle, we’d expect a mountain of debt by the time they’re on their deathbed. The obvious objection is that they’re repaying their loans so as not to accrue infinite interest. But you’re in a bit of a bind here. The older loan cannot be paid through an asset sale because that’s the very gain you’re trying to evade. If instead you refinance or roll the debt forward, then interest will accumulate; thus, we’d expect a mountain of debt waiting to greet the mountain of stepped-up assets at the point of inheritance. Empirically, this is completely unfounded.

Liscow and Fox note that of the top-1% of households aged 35-44, debt was ~8% of wealth, and it was far lower for people aged 65-74, and ~0.8% for those 75 and older. So ironically, the very thing the “buy, borrow, die” hypothesis predicts is found inverted in the empirics.

But let’s keep going down the chain of reasoning here. Maybe, just maybe, all these sophisticated tax evaders became terrified of debt the moment they qualified for Medicare (at 65 years old). Or more plausibly, maybe the average wealthy household isn’t exploiting some elaborate “buy, borrow, die” strategy. The average millionaire, billionaire & trillionaire doesn’t need to borrow against their assets to fund their lifestyle.

The Rich Simply Don’t Need to Borrow, They Just Need to Save

The entire “buy, borrow, die” myth depends on the assumption that the rich only look rich on paper, but their untaxable fortunes can’t be used to fund their lifestyle without triggering an IRS cash grab. This might be true for a minority of the rich whose wealth is locked away in illiquid stocks, but not for the average top 1%er, since their taxable base already gives them a deep pool of cash to source from.

The top 1% already is getting tax on their salaries, wages, dividends, interest, and realized capital gains. These people aren’t desperately scrounging for cash to buy some more jet fuel; they already have more than enough taxable income to fund their consumption, with only 15% of the top 1% consuming more than their taxable income.

The actual dominant strategy Liscow & Fox point out is that the rich “buy, SAVE, and then die”. Rich households spend only a fraction of their taxable income, tucking the remainder into assets that appreciate. These investments continually compound and appreciate, and the gains are then transferred on a stepped-up basis.

Conclusions & Implications: Tax Myths Persist, But We Can Change This

Throughout my blogging ‘career’, I’ve debunked an endless amount of pervasive myths, and it’s likely they’ll keep coming! The patient zero outbreak always starts something like: someone “discovers” a tidy and intuitive mechanism -> they parrot it online to their giant audience -> the mechanism is shown to exist -> everyone conflates the existence of the mechanism as proof of its frequent use.

And as I touched on earlier, scary headlines are memorable. Elon Musk borrowing $1 billion from SpaceX certainly is more interesting than some boring NBER paper. So I suppose it’s no surprise that myths like these persist.

The unspoken harm in all of this is that misinformation obfuscates genuinely good tax policy as people mindlessly grope towards stuff a Twitch stream told them to. It might actually be better to have some mix of a progressive consumption tax, land value tax, and small inheritance tax (Taiwan paper), but these are never talked about because people are too busy focusing on a fictional tax avoidance story. The empirical reality undoubtedly proves that we need tax reforms, but we can’t do it without stopping persistent tax mythology.

Speaker List:

Hamsters (me/host): https://x.com/sigmahamster2

Dabchick (co-host): https://x.com/dab_chick

LiquidZulu: https://x.com/liquid2ulu

Sinthe: https://x.com/SinthePosting

AnCap Doggo (I don’t even think he spoke once): https://x.com/Ancap_doggo

The Modern Conservative

James Fishback has optimized his campaign for Gen Z, the new wave of young voters. Quite recently, he was embarrassingly turned down after trying to catch an endorsement from IShowSpeed. This was just months after his event, where he called for the lynching of a Black onlooker who had some scandalous questions for him about his alleged relationship with a minor.

Fishback simultaneously postures as a sort of anti-leftist businessman fighting for the working-class Floridian against foreign labor, multinational corporations, Wall Street investors, and tourists? Oops, accidentally adopted every leftist position in the book, Mr. Fishback.

In any case, Fishback is an instructive example of someone whose entire campaign is tailored around whatever fragile trend hit TikTok this morning. His branding leans heavily on rattling off populist policies that appeal only to the most uninformed Floridians. It’s built on the little rage-bait internet theater that has everyone debating the implications of what he said on Twitter for days on end.

And at the end of the day, Fishback is the perfect example of the modern Conservative. In a January interview with Tucker Carlson, he stated that “Our North Star should not be a free market, but a free people.”

What Fishback really means is that anyone who can obtain citizenship ought to be granted his ever-so-gracious paternalism. He wants to selectively remove property taxes for select individuals, tax out-of-state home buyers, punish high-skill immigration (H1-B), ban data centers in all 67 counties, and ban institutional investors from buying homes.

Reagan-era conservatives mounted substantive critiques on how allocating resources based on political whims could never replace millions of decentralized choices, and on how planners could only helplessly scramble to respond to the qualitatively diverse interests of the people. Even bare-bones paternalism was criticized harshly by Reagan-era conservatives. As Ronald Reagan said, “Instead of protectionism, we should call it destructionism”.

I suppose those days are long gone, since what we have today looks like a malformed version of Hamilton. While Fishback is obviously not a ‘Communist’, it’s clear that he’s perfectly comfortable extending tidbits of central planning in places where earlier conservatives would never have. In essence, he wants politicians like himself to decide whether an employer should really hire a foreign worker, whether a man owns too many homes, or whether a university should be able to accept foreign students.

Trade & Protectionism

Like most modern conservatives, Fishback has supported the tariffs as a means to reshore manufacturing, stop job offshoring, and spur American production. In an interview in April, Fishback expressed support for Trump’s anti-globalization agenda and argued that Americans would have to choose between cheap imports and blue-collar jobs. This kind of unsophisticated industrial policy never actually bears much fruit, as I’ve noted in another Substack. The issue is that plans for reshoring never actually correspond to a blueprint for reshoring clusters of the economy regulated by hard budget constraints and performance targets, as industrial policy experts like Dwight Perkins suggest.

And like Trump, Fishback has layered plenty of sugarcoating on his rhetoric, using phrases like “fairer trade” or “reversing the hollowing out of American industry”. It makes perfect sense if you think about it. Reversing sounds like filling a major industrial gap when the reality is that a ‘reversal’ just means a guy standing at a port tacking on a tariff fee, praying that the cost burden will entice multinationals to sink capital domestically. This is rarely empirically borne out since the importer & consumer end up absorbing most of the damage. The USITC found that American importers bore almost the entire tariff, while only a few protected industries gained.

It’s not like a Florida governor can tack on federal tariffs anyway, so this is mostly another performative gesture in his campaign. Still, it’s fascinating how easily people eat up his empty promises.

Labor Immigration

Perhaps the worst part about Fishback’s entire platform is his strange views on immigration and labor. To preface, it seems abundantly clear to me that restricting immigration is just another form of protectionism, i.e., constraining the supply of labor to increase the bargaining power of the domestic population. This comes with a multitude of spillovers & unintended effects that, in my view, mean immigration restrictions are welfare-reducing.

On multiple occasions, Fishback has advocated stripping H-1B holders (high-skilled foreign labor) of their access to American institutions. His core promise is that companies that ‘displace’ qualified workers will be charged a recurring fine of ~$200,000. Functionally, this is DEI for domestic workers and arbitrary punishment for firms simply trying to survive a highly competitive business environment. Not very right-wing of you, Mr. Fishback.

To be fair to Fishback, I think the strongest paper “in favor” of his concerns comes from Borjas (2026), which finds that H-1B workers earn ~15% less than equivalent natives. Their model also predicts that deterrence in the form of a one-time fee is unlikely to change much. So, if you know H-1B immigrants will ask for less than the domestic population, then it seems intuitive that immigrants will win out at the expense of domestic workers. As Gelber and Isen (2014) note, firms do internally crowd out domestic workers by hiring high-skill foreign workers.

But remember, the core premise of this argument is that ‘H-1B workers are cheap substitutes for Americans; therefore, the state must prevent this transaction’. The argument is not “well, firms who hire H-1 B crowd people out”; it’s that the sector will see an increase in employment overall, relative to what it would have been had we not had H-1B visas.

There’s an overwhelming body of evidence against Fishback & the conservative narrative of immigration job substitution. Glennon (2020) finds that H-1B restrictions actually cause corporations to expand offshore employment, especially in research-intensive work. She estimates that for every rejected visa, a multinational hires 0.4-0.9 more workers abroad. So the hilarious irony of it all is that Fishback’s own immigration restriction evidently has the opposite effect he promises. Turns out, right-wing populism isn’t a good substitute for econometric analysis.

Peri (2014) finds that H1-B workers do not displace domestic workers, specifically in the ‘computer sector’; in fact, the opposite is true. They end up serving as a complement, which means they increase native worker productivity partly since the number of jobs was never fixed.

Education Immigration

His views on international students just might be even more AIDs. Fishback wants to charge every foreign student at a public university $1 million per year, which would apply to students enrolled in programs. Meaning, if you’re a third-year foreign student at the University of Florida, you better get to work on that transfer application and pray your credits transfer over. Or more bluntly, “pack your bags”.

The point is, though, that foreign students aren’t subsidized by domestic Floridians. A fat chunk of them are paying out-of-state tuition, which pays a huge premium relative to the in-state tuition.

So foreign students are wheelbarrowing massive sums of cash into the country, paying for professors, housing, food, and sometimes even subsidizing the education of domestic students. NAFSA estimates that international students overall contribute $42.9 billion to the U.S. economy and have supported over 355,000 jobs in the 2024-2025 academic year. I have no clue what the chain of thought is that concludes kicking out a student who pays an absurdly high tuition would help anyone.

Housing and Investment

Perhaps the slimiest populist flourish in his campaign is his housing proposals. His current proposal would ban private equity firms from buying single-family homes, force them to divest from their current holdings within 5 years, and impose $50K transaction taxes on out-of-state buyers purchasing a vacation home. First of all, we’re hating on the rich now? The ideological irony is obvious. Seems quite consistent with leftism to me. And a $50K transaction tax? Hello, highly illiquid housing markets.

Nationally, institutional investors are a tiny, tiny fraction of single-family homeowners, hovering at ~1-3% by 2024.

The caveat here is that institutional investments can be more concentrated in Sun Belt regions like Jacksonville, where they own ~22% of single-family rental homes. But whether this is an issue is a different question. Single-family rentals can be substituted for duplexes, townhouses, starter homes, large apartments, etc. Furthermore, Jacksonville is a subset of Florida, meaning there exist substitutes outside of the main city area. And if this truly was a problem, then the solution wouldn’t be to ban institutional investors. The solution would more likely be to simply build more homes through deregulating housing & zoning.

The most sincere question here would not be whether institutional investors hurt some part of the consumer market, but rather how they overall affect prices, construction, rental supply, and ease of access. (Francke et al., 2025) for example, exploit a 2022 policy in the Netherlands that dropped investor purchases by ~19% while first-time home purchases increased by ~13%. The ban seemed to have some efficacy in shuffling which demographic lived in homes, but it did not lower the price of homes, and most of the landlords affected were small investors rather than the institutional US moguls like Blackstone. Functionally, this is just gentrification for first-time homebuyers.

Coven (2025) uses a structural model and finds pretty similar mixed tradeoffs. Institutional investors increased prices and reduced the number of homes available for occupancy; however, economies-of-scale investing increased the rental supply, thus lowering rents. To quantify this, for every home an institution bought, rentals rose by roughly 0.5, and homes available for owner-occupancy only fell by 0.22. So, in short, banning institutional landlords raises rents, with smaller landlords competing for the divested housing stock.

Other dynamic models (Wang & Zhai, 2026) that study the Atlanta market found that institutional behemoths, which owned ~3% of the metropolitan housing stock, raised the renter share (% of households that rent) by 1.58% relative to the no-institutional-investor counterfactual, increasing renter welfare by $50.17. Again, very mixed, you’re mostly shuffling around different parties (crude redistribution) and exploiting any minuscule gains to motivate policies that might sound politically appealing.

You might be starting to understand why banning institutional investors seems like such an unserious policy proposal. Fishback’s forced institutional divestment plan would likely throw some extra homes into the hands of homebuyers, but it would also cripple units created out of economies of scale and shift properties to smaller landlords.

The same can be said for the $50K out-of-state vacation home tax. In the short run, out-of-state demand can certainly bid up prices; however, in the long run, we’d expect supply to increase to meet demand, unless bureaucracy hampers supply.

It’s unlikely Fishback’s housing proposals will do much, if anything, so to put it bluntly, any serious housing proposal starts with deregulation of construction and repealing zoning laws. If Florida wants cheaper housing, the long-lasting solution is to allow for less bureaucratic construction and infrastructure expansion.

Conclusions and Implications

In short, Fishback’s rhetoric and campaign are a form of deep political paternalism dressed in conservative vibes. His proposals would empower politicians to control who companies hire, who universities can admit, which institutions can purchase homes, and where people can invest. Thankfully, the probability of him winning is extremely low, but I think the fact that a campaign can be driven purely based on viral ragebait and populist rhetoric is a serious concern.

Further Reading

I have far more radical views on labor immigration, but they are not so relevant to the topic at hand, so check out this other blog if you’re interested in my fleshed-out view.

Singapore is often cited as a case study by everyone on the political spectrum. I mean, who wouldn’t want to take credit for an overcrowded backwater that transformed into a spotless, wealthy metropolitan city? Changi Airport feels like a luxury mall reserved for the elite, and everyone knows about the luxury boat hotel. Best of all, there are countless recordings of former Prime Minister Lee Kuan Yew giving powerful & inspirational speeches about having an iron spirit.

And for Libertarians in particular, there are some genuine laissez-faire aspects to Singapore. The corporate tax rate is ~17%, imports can dock at ports with mostly tariff-free access, and foreign investors are treated quite well. Of course, this is only a sliver of the broader economy.

I recently asked my followers on Twitter why laissez-faire Libertarians & Anarcho-Capitalists are so obsessed with Singapore when it’s very far off from their proposed model.

The responses were unsatisfactory to say the least.

Everyone here is, of course, not wrong in pointing out the obvious low taxes, free trade, low barriers to forming a business, strong private property laws, and Singapore’s ranking on most economic freedom indexes; however, these are incredibly cherry-picked facts that don’t give us a sufficient picture of Singapore’s overall economy.

Singapore is a heavy-handed developmental state that publicly owns the majority of land that provides housing for the majority of the population. Singapore also mandates savings and plans for industrial production, infrastructure, and defense. Calling this arrangement “laissez-faire” seems way off.

For this blog, I’m willing to exclude Singapore’s social authoritarianism since we’re only interested in economic freedom & laissez-faire economics, though, as a quick reminder, Singapore has judicial caning (spanking), along with other radically harsh punishments such as the death penalty for drug traffickers who carry more than the statutory threshold (15 grams of heroin or 30 grams of cocaine). In any case, though, I think I have more than enough substance to demonstrate the non-libertarian nature of Singapore’s economy.

Singapore is spectacularly successful, hence why so many people want to reverse-engineer their ideology into it as a success case. Socialists do this, and Libertarians do this. “The country is rich, therefore it must be laissez-faire (or socialist). The reality is that Singapore is a market-friendly developmental state with a highly activist but benevolent government.

State Ownership of Land

Singapore’s Housing and Development Board (HDB) has built over 1.25 million flats, which make up roughly 80% of residential lodging. In 2025, 77.2% of residents lived in HDB dwellings alone. These HDB households are labeled as homeowners, which is much different from being a homeowner in the United States. Similar to China, you’re merely purchasing a long-term lease with interest rather than gaining ownership to appreciate like an alternative asset. Long-term lease arrangements stretch to 99 years, and selling usually requires a minimum occupancy of 5 years. So, to be crystal clear here, the state floods the housing market, sets the rules, designs mortgages, and determines eligibility? Not very libertarian, I must say.

Properties aside, ~90% of Singapore’s land is owned by the state. There’s not an ounce of homesteading here. The Singaporean government did this by inheriting colonial Crown colony land and using compulsory acquisition laws to retrieve the rest of the island. There are genuine fruits borne out of this, too; Singapore transformed swathes of slums into dense, transit-connected cities and made housing extremely accessible. On top of all this, Singapore’s housing system is tied to a mandatory social security machine funded by employer and employee taxes. In 2026, the rate for citizens and permanent residents under 55 was 37%, with 20% coming from employees.

You can argue how this YIMBYist setup is far more optimal compared to the mixed American model of zoning and local scarcity cartels, but still, this very clearly is not the poster child of a laissez-faire libertarian model.

The Sovereign Wealth Funds (Temasek) and Capital Allocation

The cherry on top is Singapore’s Sovereign Wealth Fund (SWF), Temasek, which looks like something straight out of Econoboi’s wet dream. The colossal fund of 518 Singaporean billion dollars (401.06 billion USD) invests heavily in domestic companies like DBS, Singapore Airlines, Mapletree, and so on. Similar to South Korea, Singapore’s industrial policy used government-procured funds to carry a cluster of industries from finance to aviation to electricity.

The government also offers various incentives like tax credits and grants for certain projects, committing roughly 37 billion Singaporean dollars to semiconductors, AI, and biotech over 5 years.

Neither of which is central planning or libertarian Capitalism. It would be appropriate to call this kind of setup a polished Capitalist activist economy where bureaucracy isn’t overtly hampering everything down.

The Problem With Economic Freedom Indexes

I believe the deepest confusion lies in the use of economic-freedom indexes. Essentially, indexes are designed by institutions to measure freedom, not to measure how close an economy is to the preferred Austrian model. Could you imagine that? An index in which 100/100 meant an Austrian AnCap economy, 75/100 meant a Chicago School-esque liberal society, and 50/100 meant some Dirigiste command economy. Unfortunately, that’s not at all what indexes approximate. Unironically, when AnCaps talk about freedom index scores, they talk as if someone’s piping around a jar of freedom to pour on select countries.

Anyhow, if you look at any Freedom Index, you’ll likely find Singapore at the top. The Heritage Foundation, for example, scores Singapore at 84.4/100, ranking #1. To quickly break down this score, their tax burden score is 89.2, government spending 58.3, business freedom 89.5, trade 95, and investment freedom 90. Not having to sacrifice your left ball to a licensing bureau every time you want to take your company public is certainly a plus, and pretty closely aligned with your typical laissez-faire model.

Heritage gives each of its twelve components equal weighting, i.e., government integrity counts for 8.33%, trade freedom counts for 8.33%, etc. The issue here is that there isn’t a measure of state ownership of enterprises (SOEs). And a government that owns airlines, telecommunication networks, seaport operations, and swathes and swathes of properties certainly seems more than negligible. Thus, this kind of omission is quite misleading for assessing state-directed, SOE-heavy countries like Singapore. It would be like observing an economy with no export barriers, while ignoring that the very building and ground that the company operates on is government-owned. The Fraser Institute mends some of my objections here by including a measure of SOEs, but heavily underweights them in the final calculation, and includes measurements like sound money, which are wholly irrelevant to the question of whether a country’s economy is laissez-faire or not.

Other Index Problems

To jump back to the Index problems, the Fraser Institute does a better job than the Heritage Foundation by adding in a measure of state ownership, by which Singapore scores 5.145/10, a clear acknowledgment of anti-libertarianism. Unfortunately, the weighting is indefensibly low, sitting at 4%, which means that Singapore’s heavy SOE presence only drops the total freedom index by (10-5.145)*0.04 = 0.1942%. Perhaps this is defensible for measuring how free people feel or something, but laundering this into measuring how close the economy is to a laissez-faire setup, it becomes indefensible. It’s absurd to think that the state owning ~90% of the land, dominating the housing market, forcing savings, and holding major stakes in corporations across the commanding heights of the economy would be merely a drop in the bucket.

But to reiterate, Singapore’s ranking is not fake, more than it’s skewed by irrelevant metrics. Singapore really is remarkably laissez-faire on free trade, taxation, foreign investment, etc. Interestingly enough, tinkering around with the SOE weights can show you how sensitive the scale is.

And if you tweak Singapore’s 5.145/10 SOE score and change the weight from 4 to 10 percent while lowering everything else proportionally, then Singapore drops from rank #2 to rank #4. At 15%, Singapore falls to fifth, and this is without tweaking every other country on the list. Tweak the specs a bit more, and at a 20% weighting, Singapore falls to fifth. You can validate this yourself by downloading the Excel sheet on Fraser’s website.

But is tripling the state ownership weight really defensible? I think so. The rationale I have here is that giving state ownership a 15% share of the index is justified by a 5% share for nonfinancial state enterprises, 5% for land ownership and housing, and 5% for mandatory savings & preferential credit handouts. So in essence, I believe the design choice, at least for the question of laissez-faire economic models, is incorrect; thus, these indices are not informative of how laissez-faire an economy is.

Corruption & Competence

The other glaring issue is one you’ve probably heard of. These indices always mix institutional setups with successful outcomes or qualities unrelated to the setup. Less corruption often means more freedom, yes, but Libertarians don’t suddenly get to bake in less corruption into their minarchist economic model in virtue of this. The same thing applies to crime rates, police effectiveness, labor force participation, GDP growth, etc which fail to tell us how closely aligned this country’s economic model is to our own. If you launder this into your index and cite it as proof of laissez-faire, then what you get is a circular exercise where prosperity is baked into our definition of freedom.

Anyhow, I couldn’t reconstruct the index with the Heritage Foundation’s index because their twelve component scores don’t have a publicly available breakdown, quite strange, I must say. But the idea is simply to exclude all these variables that are irrelevant to the institutional setup. With the Fraser Institute, I removed police, crime, current inflation, inflation variability, and money growth measures, which make up roughly 17.5% of the original index. By renormalizing the “survivor” metrics, I adjusted Singapore’s score to 8.53. Then, by readjusting the SOE weights (15%) as done in the previous section and scaling it up by 20.48%, we end up with a new SOE weight of 18.72% or ≈ 19%. Based on this metric, Singapore now scores between 5th and 8th place.

Some Caveats

More importantly, Singapore is undoubtedly one of the best industrial development stories in the past century, but the counterfactual is not exactly clear. We only observe one Singapore, and there isn’t some clean control-treatment split we can peer our eyes into. As political economist Bryan Cheang pointed out, Singapore’s success is real; however, Singapore could have been far more successful if it had pursued a more privatized-free-market setup, in which private firms scale into global competitiveness. Cheang points out that directing land, capital, education, etc, towards favored multinational firms undercut smaller entrepreneurs as they were swallowed up by the uneven playing field. Pre-1997 Hong Kong is a case study which makes a strong case for liberalized development. In any case, something to keep in mind.

Conclusions & Implications

The important thing to keep in mind here is that Singapore is a Capitalist economy and central planning likely has little to do with its success; rather, Singapore is a polished state-led economy with light regulation on ordinary business activities.

Libertarians should certainly point to Singapore’s low-regulation environment, but they shouldn’t ignore the public land ownership, compulsory saving, and swathes of public housing that make Singapore very far off from their ideal.

Introduction

It pains me to say this, but the Democratic Socialists of America should be taken seriously. In 2016, Bernie Sanders nearly won the Democrat primary. Just a decade later, Alexandria Ocasio-Cortez has garnered a massive national following while injecting fringe leftist ideas into the mainstream.

Spend a few minutes on the DSA’s website, and you’ll encounter a utopian vision sprinkled in with populist flourishes that are utterly detached from contemporary economics.

For one, they persistently conflate “for-profit” private enterprise with socially undesirable outcomes. This is despite Capitalism having lifted millions out of the depths of bare-bones feudal subsistence.

The deliberate confusion of means with ends isn’t a mistake, though. It’s a common rhetorical trick aimed at appealing to the most disenfranchised Americans who want to compress all their problems into one word: profit.

As Milton Friedman once observed, “One of the great mistakes is to judge policies and programs by their intentions rather than their results.” Alas, shaking your shiny populist messaging of “good intentions” is often more salient to voters than a monotone lesson on economics and tradeoffs.

The shiniest new toy the DSA has to offer is the promise of a 32-hour work week. They envision a future in which everyone gets an extra day off to bask in the sun on a beautiful socialist Sabbath. You can practically hear the populist pitch: “Well, wouldn’t you prefer to sleep in on Mondays and then water paint in the afternoon?” The rational answer is a resounding “yes, of course.” Did I mention I would also like an all-inclusive trip to South Korea, a Cybertruck & a government-granted girlfriend?

In any case, the harder question the utopian socialist never contends with is the cost-benefit analysis of such a radical mandate. If you mandate fewer hours while promising to keep pay and benefits, something must give. Given the vapid demonization of private enterprise, it’s plausible to believe that most DSA members simply expect businesses to quietly absorb the increased labor cost. But even if businesses did do that (they won’t), what kind of implications does that have for the long-run productive capacity of the economy? Surely the long-run output will suffer if we deliberately incentivize lazing around.*

If Crusoe and Friday are stranded on a desert island, is it really wise for them to expend labor as little as possible and live on bare subsistence? Or would it be better for them to work harder if it confers more benefits down the line?

I believe the latter is preferable; there’s something uniquely exceptional about the desire to chase more, to build more, to pursue ambitions, even when our basic needs have been satiated.

Now imagine if Crusoe and Friday were told by some paternalistic sky daddy to take more time off, to put off the boat building, to just spear fish and sip on coconuts for another day. At the very least, their ambitious spirits shouldn’t be thwarted by paternalistic bureaucrats hellbent on imposing leisure. The opposition seems to disagree.

The most revealing example of this comes from AOC in 2020, where she discovered that famed economist “Milton Keynes” had predicted that by 2030, industrialized capitalist economies would grow so much that the workweek would be just 15 hours.

Economic Possibilities for Our Grandchildren

In John Maynard (not Milton) Keynes’ 1930 essay “The Economic Possibilities for Our Grandchildren”, Keynes predicted that economic output and living standards in industrializing capitalist countries could leap four to eightfold over the next century, leading to a 15-hour work week. Considering this prediction was made during the depths of the Great Depression, it was considered abnormally optimistic. However, nearly a century later, a BLS paper noted that long-run economic growth (GDP) had fallen in the range of Keynes’ forecast, while his forecast of a 15-hour work week was far off.

The standard explanation offered by people like AOC, Sanders, and the DSA is that capitalist economies mysteriously funneled most of the productivity gains to the top 1%, leaving everyone else with stagnant living standards despite the increase in output. Funnily enough, leftists, especially Marxists, use this as a way of litigating Keynes’ Capitalist naivety. But this is an incomplete and incredibly uninformed explanation.

My reading of this is that Keynes made a profound error in projecting his own preferences onto everyone else. Similarly, the DSA acts as if there are some kind of universal preferences rather than individual ones. How socialist of them.

So despite enormous productivity gains in the past century, the share between leisure and higher material consumption leaned overwhelmingly in favor of the latter. Working hours, of course, did decline from the brutal early industrial schedules, but people also chose to spend their higher incomes on larger homes, higher quality healthcare, better housing amenities, and countless other innovations that Keynes could not have fathomed.

Imagine telling Keynes about a pocket supercomputer that exponentially outpaced anything from his era in terms of computational speed, storage, memory, etc. The point is, though, productivity gains were not funneled exclusively into leisure or captured exclusively by the top 1%. They appeared in everyday walks of life, from higher-quality goods to higher quantity consumption.

How Capitalism Deals with Leisure

There’s another mistaken view layered beneath the DSA’s platform, too. The mistake is in thinking that leisure and consumption are distinct. Leisure in itself should be thought of as a form of consumption, i.e., every hour spent away from the office is an opportunity cost.

People naturally have different preferences: investment bankers are happy to sacrifice weekends for a chance at Goldman’s executive branch, while newlywed moms are vying to land a three-day hybrid work schedule. So it’s hard to accept any universal mandate when you consider the qualitative differences and choices between leisure and labor.

How The DSA Deals with Leisure

The tamest version of this proposal comes from Bernie Sanders, who promised to lower the overtime threshold from 40 to 32 hours a week. So if Bernie’s master plan is to ever come to fruition, any hour beyond this 32 would be paid 1.5 times the standard rate, with the expectation that employers would normalize shorter workweeks. Because of the novelty of such a proposal, I can’t point to particularly strong empirical evidence or robust research designs that allow us to quantify the long-run effects of such a policy, but a few numerical examples can illustrate the obvious economic factors that someone will have to bear.

Suppose a worker earns $1000 for a forty-hour work week. Thus, the employer’s hourly labor cost is $25 an hour. Now, imagine if the employee continues to work 40 hours a week after the 32-hour overtime bill has been passed. Suddenly, the weekly labor cost is $1100: $800 for the first 32 hours and $300 for the remaining 8 OT hours.

Considering most competitive markets live off razor-thin profit margins, the cost will undoubtedly shift onto consumers through higher prices or onto employees through layoffs.

But there’s also a clear workaround for employers. Employers may end up restructuring compensation so that the 32 hours paid at the standard rate plus the 8 hours paid at the overtime rate cost the same as before the 32-hour OT mandate. And if the bill attempts to ban pay reductions to affected workers, then we’ll likely see layoffs or stagnant wages.

Why The 32 Hour Work Week is Evil

I believe there’s a strong theoretical case (from abduction) as to why the 32-hour workweek is bad economics. But I think the broader critique here is that this kind of policy is deeply paternalistic. If millions of people in the pursuit of a perpetually better life are to have their preferences forcibly superseded by a handful of bureaucrats, then I’m not sure this is anything less than evil. By using the law to privilege leisure over labor, you overwrite the preferences of millions of ambitious Americans. Yes, some people simply want more free time, but others want to build, others want career advancement, and others want to save more. A universal mandate treats qualitatively different ambitions as one neat pile of clay ready to be molded by self-anointed social engineers.

“The claims of these organizers of humanity raise another question which I have often asked them and which, so far as I know, they have never answered: If the natural tendencies of mankind are so bad that it is not safe to permit people to be free, how is it that the tendencies of these organizers are always good? Do not the legislators and their appointed agents also belong to the human race? Or do they believe that they themselves are made of a finer clay than the rest of mankind?”

— Frederic Bastiat, The Law

Footnotes

* I am aware that there is some evidence of long-run output increasing in response to labor hour reductions, but those are typically sector-specific, and if there are productivity gains to additional leisure hours, then private firms would recognize this and enact it.

https://ideas.repec.org/a/bla/indres/v62y2023i3p217-232.html

“In this paper, we exploit a panel of industry‐level data in European countries to study the economic impact of national reductions in usual weekly working hours between 1995 and 2007. Our identification strategy relies on the five national reforms that took place over this period and on initial differences across sectors in the share of workers exposed to the reforms. On average, the number of hours worked in more affected sectors fell, hourly wages rose, but employment did not increase. The effect on value added per hour worked appears to be positive but non‐significant.”

Donald Trump has been very explicit about his intention with tariffs. He wants to make offshoring factories expensive enough that building a factory in the Rust Belt of America doesn’t look too bad by comparison. In Trump’s heroic story, tariffs are the disciplinary tool meant to whip American industry back into shape after decades of “hollowing out”.

The unfortunate part is that the strongest critique of Trump’s tariff gambit is left out of all mainstream political talks. Democrats have mostly spent their TV airtime reaching for the laziest critique, that is, “tariffs are bad because part of the tax incidence falls on consumers, and no one could ever hurt the wittle old consumer at the grocery checkout”, which obfuscates the bigger issue. The uglier problem is that Trump’s “industrial policy” misses virtually every strategic checkbox in the industrial policy literature. Trump’s attempt to revitalize American industry lacks the essential milestones, public investment, and long-term regime continuity that make industrial policy attempts like South Korea’s workable.

I could spend the rest of this blog post doing victory lap propaganda for free trade; throw in that Milton Friedman quote calling for unilateral free trade, cite 1960s Hong Kong a few times, and list out the dozens of failed examples that preceded South Korea. However, I think the more interesting point here is that Trump’s plan for industrial policy fails even on its own terms. The long story short is that industrial targets, export discipline, and not getting caught by a sunk-cost fallacy are all incredibly important to making this work. Trump has failed on all of them.

I want to quickly provide some much-needed South Korea history to show why South Korea’s modern economy is so impressive. In the 1950s, Korea was a war-torn backwater in which ~60% of residents lived in extreme poverty. Before that (1910-1945), Japan had colonized Korea. Before that, Korea (Goryeo) was repeatedly subjugated by the Mongols. Now they’re a modern metropolis after decades and decades of aggressive industrial policy, but Trump’s plan won’t end the same.

In Chapter 3 of Dwight Perkins’ book “East Asian Development: Foundations and Strategies”, he details exactly why South Korea’s industrial policy was so successful. The East Asian model essentially started by picking strategic sectors, i.e., steel, shipbuilding, electronics, machinery, petrochemicals, and nonferrous metals, which are all pretty connected in terms of skillset. This doubly means that they’re complementary to one another as they expand towards economies of scale. The whole point is to reach economies of scale where you’re producing everything at a lower cost than your global competitors, and to be linked to other important industries that will give your country a plausible pathway to international stardom. When you look at US attempts, reindustrialization attempts, they usually just “strategically” target a single industry, like steel, and call it a day. So it should be no surprise to find out that the USITC estimated that Trump tariffs between 2018 and 2021 raised steel production 1.9% ($1.3 billion), at an opportunity cost of $3.5 billion for the steel-aluminum industries.

South Korea, for example, didn’t throw cash into “heavy industry” and call it a day. It picked six sectors with high complementary spillovers and successfully made them globally competitive. To grasp the problem that industrial policy is fixing, imagine an empty bay.

You could add a massive shipyard here, but private investors are completely unwilling to finance this since there aren’t enough steel mills or machinery to make this investment lucrative. The industrial policy fanatic is looking to build the entire cluster of goods: the steel, the rolling mill, the blast furnace, the ship, and with it, the entire economy. So it’s mostly a coordination/market failure argument that “industrial policyists” make. The important provision here is that South Korea coupled all of this with specific export targets and international competitiveness litmus tests. Trump has not only failed to enact such milestones, but he’s failed on the prerequisite to this step, too!

The other issue is staving off lobbyists and staying incredibly disciplined. Lobbyists are naturally self-interested and thus are not interested in what’s good for the broader economy. Favored elites (Chaebols) who constantly get flushed with public funds certainly exist, but their power accumulation at least was dependent on meeting industrial policy export targets.

Now contrast that with Trump’s Intel deal. The admin pushed $5.7 billion that was meant to be conditional funding under Biden’s CHIPS Act. The admin subsequently slashed all existing project milestones, like disbursement requirements and non-activist equity stakes. They essentially skipped the most important part of disciplining firms. In other words, they fitted their holy knight with armor and sent him into battle with zero expectations or goals.

The second component I’d like to highlight is soft bank loans, i.e., subsidized and direct financing. All of these industrial companies need to sink massive amounts of capital today in the hopes of a much bigger payoff way down the line. South Korea’s government pushed for subsidized loans on a quid pro quo basis towards these prioritized industries. The quid pro quo part is a shorthand for saying that it was a revolving credit line that had internally competitive prerequisites. So the goal was soft bank loans, but a hard line budget constraint where losers aren’t perpetually bailed out. Cheap money is only afforded to those with strong export growth, productivity gains, or technological progress.

The worry with the Trump administration is that if you allow for a soft budget constraint where losers are perpetually bailed out, you have less of an industrial policy and more of a corporate welfare ATM. Trump’s Intel deal should have you strongly updating in favor of the hypothesis that he’s utterly incapable of leading disciplined industrial policy. Under Trump, Washington repurposed $8.9 billion in CHIPS Act funds to purchase a 9.9% stake in Intel, a company that was not performing well whatsoever. As a fair caveat to this, Trump and the DoD did cite this purchase as a national security hedge, but I’m reading this more as a hedge for their industrial policy failures.

If you analyze South Korea’s industrial policy success, you’ll notice that missing targets meant subsidized credit deals soured. Firms that were juiced up with leveraged debt desperately chased this, creating a strong incentive to meet targets. There’s no backdoor where you can just say, “Oh well, we actually underperformed this quarter because we didn’t have enough funding, so give us more funding”. Keep missing targets, and you’ll lose protectionist tariffs, get your management internally reshuffled, or just have your company slowly wilt away. Authorities were notoriously tight on monitoring the progress of favored firms, ready to pull the plug on a moment’s notice (though this is less true in modern terms with Chaebols and corporate capture). The point is that if companies believe that Seoul or Washington will bail them out at every critical turning point, then any targets become flimsy suggestions. Without a strong disincentive or punishment for failure, global competitiveness is just a pipe dream, and subsidies are just corporate welfare.

The third point, which is not actually in Perkins’ thesis, so take it as my own adaptive thesis, is that industrial policy requires long-term regime continuity. Steel mills, chip fabs, and petrochemical complexes require far more grounding than just a four-year presidential seat. Park Chung Hee, South Korea’s former leader, who is credited with the aggressive industrial policy, did so through launching five-year plans beginning in 1962, Petrochemical industry conditional investments in 1973, and more. Korea’s 5-year planning continued all the way up to the 1990s. The quid pro quo loans and arduous industry building happened across decades, not years. And not being overturned by whatever bipolar electoral whims the people were feeling the next day. The issue with American institutions, to put it bluntly, is the highly volatile overton window and democratic process that puts friction on any type of long-term state-led planning. Industrial policy cannot survive without decades of regime continuity, or at least, policy continuity.

To illustrate how volatile American politics is, just think about the Republican Party of the 80s vs the Republican Party today. Ronald Reagan (1989) was famously a Friedmanite free trader, to the point where he called any form of protectionism “destructionism”. Meanwhile, today we have Republicans preaching the exact opposite, spitting out vague platitudes of “fair trade”, whatever that means. Democrats who were historically domestic interventionists were so infatuated with “consumer protection” that they had to oppose the very domestic industry protections they were probably secretly aligned with. Democrats have their own separate tariff garden going on with chips & fabs (CHIPS Act). More importantly, though, the constantly shifting, oppositional 2-party system of American politics means that strategic industrial targets are changing every few years, making a factory with a 15-year payoff completely infeasible. Without regime continuity, the hopes of a solid industrial policy are virtually non-existent.

Do any US politicians have the heart to pull the plug when strategic tariffs and industrial policy fail miserably? What are the actual guidelines they’ve set? Seems to be none.

Funnily enough, even inside Trump’s regime, policy continuity seems to have the lifespan of a fruit fly. His broad tariff enactments were launched under the IEEPA in April of 2025 and were shot down in just 10 months by the Supreme Court. Now they have a 10% “quasi-tariff” (customs duty) via Section 122, which probably is only going to last 150 days (July 24th, 2026) unless Congress renews.

Trump tried to copy the easiest part of industrial policy through sweeping protectionism, but failed to set benchmarks as suggested by roughly every IP economist. Without these, reindustrialization is just a euphemism for corporate welfare.

Book Citations

Joseph J. Stern, Jihong Kim, Dwight H. Perkins, and Jungho Yoo, Industrialization and the State: The Korean Heavy and Chemical Industry Drive.

https://www.amazon.com/Industrialization-State-Chemical-International-Development/dp/0674452259

East Asian Development: Foundations and Strategies (Chapter 3)

https://www.amazon.com/East-Asian-Development-Foundations-Strategies/dp/0674725301

Siberia is a fascinating region. Winter periods can see weather as harsh as -90 degrees Fahrenheit. Three human species have lived in Siberia: Homo sapiens, Neanderthals, and Denisovans. Various nomadic tribes, including the Scythians, Enets, Nenets, Huns, and Xiognu, have historically inhabited Siberia. They didn’t stay long, though. The cold seasons were often the temporary eviction notice that signaled migration.

However, under the Tsar, and later the Stalinist regime, Siberia’s harsh winter and miles of empty taiga found a greater purpose. Their vision for Siberia was one of political suppression. Send people far enough east, and they completely vanish from familial ties, politics, and newspaper headlines. Roughly 800,000 people were exiled to Siberia during the 19th century, and by the turn of the 20th century, Stalin had scaled the gulag network massively.

Being in the most remote of places, escaping the compound was only the first of many hurdles. Once you actually pass the prison fence, you’d be met with a thousand kilometers of empty taiga, mountains, and tundra. Good luck getting back to Moscow without your skin turning black from frostbite..

I almost want to throw out a fine-tuning hypothesis after reading about how perfect Siberia was for harsh prisons, since underneath all the secluded frostbitten misery was gold, coal, nickel, and timber. Ripe for rapid industrialization, Stalin exploited the region through a gulag network that shuffled around 18 million enslaved prisoners who would tirelessly mine away. It functionally was a conveyor belt feeding human toil into the Soviet industrial machine.

Prisoners mostly spent their days hacking down trees, hauling construction loads, mining ore, or whatever the planners demanded. They’d work under snow and nightfall, all under the coercion of armed soldiers. Food was also tied to performance, meaning a prisoner who failed to meet a quota received less food and thus had less strength the next day. Survivors of the gulag system described this as a sort of downward spiral in which one missed quota spelled death. This gave rise to the word “dokhodyaga” which described a prisoner who had reached the final stage of physical exhaustion and was on the brink of death. If you ignore the mountain of bodies, then yes, the economics of the Siberian system were quite good at extracting timber, ores, and other essential resources.

More specifically, though, notorious camps in Norilisk transformed into mining hubs rich with nickel, cobalt, copper, and platinum, feeding the growth of nearby cities like Magadan and areas around Dalstroy. Between 1869 and 1906, the Tsar had sent roughly 30,000 inmates to their penal colony, Sakhalin Island, giving the country another production boost. In 1890, Anton Chekhov unsurprisingly documented various abuses, from forced labor to beatings to neglected children to women being pimped out.

Famous writer Fyodor Dostoevsky got his own taste of imperial prison after getting arrested in 1849 for joining an intellectual group known as the Pertashevsky Circle. For this, he was put before a firing squad, only to find out that it was actually just one of many cruel psychological punishments the Tsar put people through. A harmless prank. He spent four years shackled in Omsk and documented the abuses and harsh conditions in his book “Notes from a Dead House”.

Ironically, some Siberian exiles later went on to scale up the inhumane systems that they had once experienced. Lenin spent three years in a relatively lax administrative exile in Shushenskoye while Stalin bounced between arrest, exile, and escapes.

Hard labor camps aside, though, it’s important to remember that Siberia is massive. ~77% of Russia is Siberia. Siberia is 13.1 million square kilometers. It’s not like Siberia needs to be maintained like a garden, but there needs to be some semblance of control, such that exiling prisoners here is useful in establishing Russian or Soviet occupancy. So much so that imperial Katorga tied four-year prison sentences to eternal settlement requirements, meaning they were forcibly sealed in Siberia. Essentially, in more economically friendly terms, Moscow subsidized the supply of miners, craftsmen, and farmers, etc., in Siberia as a regional strategic investment.

And in modern terms, Siberia’s economy is more or less sourcing from the same natural resources of timber and pulp, coal, and now natural gas. Pipelines, processing plants, and railways have made resource transportation far less strenuous on workers than centuries ago. The only catch here is that the geographic advantages still exist. Settlements are stretched thinly across the regions, connected only by permanently frosted roads. The labor market looks tightish with unemployment at 2.8% in 2024 (note this is a Russian site, so I’d recommend clicking it with caution). The biggest weakness in Siberia’s modern economy is that fixed cost infrastructure projects are fat since the cost is spread across very few individuals.

And actually, interestingly enough, the biggest & most underrated strength of the Siberian economy is its political value. If you send a prisoner thousands of miles east, the seclusion acts as a sort of speech suppression. Families now need to spend on airfare, trains, hotels, and permission from bureaucrats just to visit in person. Even lawyers end up hearing radio silence with their clients after some microscopic, highly malleable prison “violation” occurs. Journalists, too, are left out. More importantly, Amnesty International has documented various interactions where Russian authorities have arbitrarily restricted contact with families and cancelled visits on a whim. The remoteness of Siberia does half the censoring for free, and the politicians can make up the difference.

Alexei Navalny, for example, disappeared somewhere in the Russian prison system in December of 2023, lost contact with his lawyers on December 6th, and then missed his future court hearings. Not a single explanation by the state or a pinpoint location, just a shrug while his family desperately searched. After 19 days of radio silence, they found him at “Polar Wolf”, a penal colony in Kharp, which was 1900 kilometers northeast of Moscow. Is Navalny’s disappearance within the Arctic prison system just coincidentally aligned with Putin’s reelection campaign? Seems more plausible to say no.

Ultimately, modern Siberia doesn’t need to perfectly replicate Gulag networks to preserve the political value Stalin and the Tsar enjoyed. Siberia has gone through plenty of phases, from being a nomadic frontier to a penal colony to an industrial powerhouse. So Siberia has never actually been just empty tundra; it’s a resource-rich mining hub that gives politicians like Putin control.

Scott Sumner’s Socials

Substack: https://substack.com/@ssumner

Hosts & Guests

Hamsters (me): https://x.com/sigmahamster2

Sal: https://x.com/GirlBossGuru

Tyrus: https://x.com/tyrus421

Benja: n/a

Whatever happened to fun careers like being a knight? Or a samurai? Or essentially any warrior role? The short answer is that economic advancement & state capacity killed them, and with that, their prestige slowly rotted into the societal ether. Turns out you don’t need a long lineage of bloodthirsty berserkers to assert authority.

The first millennium’s “use-value” of old warrior classes stemmed from a brutal, practical need. You need men with swords. You need experts in violence who can crush rival factions, defend your plot of dirt, scare off bandits, establish order, and show power beyond mere bluff and abstraction. This violent authority, comprised of thousands of trained warriors, was the budding stage of the state.

In Hollywood, you’ll see European knights romanticized as shiny, armored heroes, galloping through misty fog on a dirt battlefield towards a castle moat or whatever. There’s, of course, a sliver of truth to all this, and cavalry was no doubt an integral part of a good defense; however, by the time economic progress began to ramp up, their use-value was next to none. Times of peace meant resources could be dedicated to productive innovation rather than configuring your economy to a stagnant Keynesian war economy where half your GDP is put towards war infrastructure. Gunpowder especially changed the landscape. One amateur marksman could now execute people from a safe distance, functionally serving the purpose of several knights who spent their entire livelihood training, that is, years of opportunity costs. What’s scarier than a guy with a sword? Probably a tax ministry that can tax three villages, conscript 1,000 men, & issue 2,000 muskets all on a moment’s notice. Oh, and it’s all under the innocent guise of public works. So, very obviously, knights were no longer an invaluable war machine, and they were forced into administrative roles. Hilariously enough, it’s sort of like textile workers getting replaced by machines; they were the aristocratic version of the Luddites.

The Samurai went through the same humiliating job automation, and like knights, they started with a genuine use-value. Samurai emerged during the Heian period, in which their role was similar to that of knights. They’d protect estates, enforce contractual claims, and restore order in the most lawless of places. At the time, violence and fighting were genuinely valuable skills, whereas today, the most value you’ll really squeeze out is giving a show at a fighting league or the UFC, and even then, only the most elite of elite can make it. In any case, by the time the Kamakura (1185-1333) period crept around, samurai had been institutionalized in the upper echelons of society. The shogunate leadership was an exclusive military club, and by the Sengoku period (1467–1600), Daimyo fights erupted, and regional warlords got shuffled around rapidly. The Samurai function was at its peak, and their social status was effectively anchored in bloodshed. But once Japan experienced a moment of peace and industrialization (Meiji Restoration/1868), the value of the warrior class was nearly worthless. Without a giant Mongol invasion to fend off, there’s no justification to burn half the budget on samurai stipends. The humiliation ritual digs a bit further than just deepening debt, though. Samurai, who were historically part of a noble caste, were quietly forced to borrow money from merchants. Merchants, aka the class that was supposed to be beneath a samurai on every level.

Weber’s (pronounced VAY-ber for some reason) point here is that modernization develops by expropriating the expropriaters, a phrase he borrowed from Marx. In other words, the local lords, shogunate, and hereditary retainers were all slowly stripped of their power and compressed into the modern administrative state.

The actual process is quite messy, though. Ideally, samurai would just smoothly transition into domain managers, tax administrators, record keepers, and some might even become police officers, which at least touched on their samurai role.

Of course, reality is never really that clean, and outdated parasitic classes will always dig their heels in and fight violently against change. Elites don’t just peacefully fork over prestige when their productive function evaporates. Economically, samurai are strapped for cash, and their stipends are fixed to mediocre rice amounts. Debt accumulates while they’re stuck in a role, slowly withering away, acting as a rentier class.

More importantly, though, notice how the prudential reasons for obeying authority changed over time. The Weberian thesis is that traditional authority commands via inheritance, sanctity, “eternal yesterday”, & enough ritualistic wrapping paper that people feel obligated to listen. You obey the shogun because they’re the freaking shogun! You obey the king because he’s the freaking king! Consistent with Weber’s thesis, the authority eventually transforms into one of “legal-bureaucratic” authority. The prudential reasons here are much different, since you don’t obey a tax collector or a government official because of their bloodline. It’s purely from the laws and regulations that govern every walk of life.

By the mid 19th century, the Meiji restoration was underway, and the parasitic nature of the samurai was abundantly clear. In a final dramatic convulsion, the samurai class launched the Satsuma Rebellion (1877), in which former samurai violently protested against the Meiji state, which had outlawed carrying blades in public, and hollowed out their military monopoly (1876 law). As Weber noticed, the state had “claim(ed) the monopoly of the legitimate use of physical force within a given territory.”

You can either serve as a cog within the new administrative machine, but you may not compete with it in any capacity. Imagine being part of a hereditary noble class for centuries, only to get kicked down below merchants in the social hierarchy..

The irony in the rebellion is that it was mostly fought with guns, so the samurai vestige of swords and spears had already been long severed. Eventually, the rebellion ran out of ammo and ended up just picking up their blades one last time; it’s poetic, really. But in any case, this kind of ‘class conflict’ failed to keep the samurai class afloat.

Weber’s initial observation on the transition between traditional authority to legal-bureacratic authority is borne out in both case studies quite cleanly. The reasons for modernization may vary, and the reasons for obeying the authority may change, but the authority remains, even if the authority rebrands itself from a sword to a machine.

Walter Block’s Socials

Substack: https://substack.com/@walterblock

Twitter: https://x.com/WalterEBlock

Hosts (Hampsters, Tyrus, Dabchick, Lucario)

Twitters:

https://x.com/sigmahamster2

https://x.com/tyrus421
https://x.com/dab_chick

https://x.com/chxseatlntic444