2025-08-14 22:51:30
Moscow is the biggest city in Europe. How come?
It’s one of the northernmost capitals in the world, yet away from the warming influence of the sea. Why?
Most European capitals are on a major river, but Moscow is on a tributary of a tributary of a river that ends in a lake.1 And while most other big cities are on river confluences, Moscow isn’t. Why such a poor location?
At over 17,000,000 km2 in surface area, Russia is by far the largest country on Earth.2 Of all the possible places it could have a capital, why did it end up being in Moscow? A city so far to the west of the country, on a tiny river, far from the sea or any trade hub?
You might retort: It’s so far west because it conquered everything to its east.
But that doesn’t really answer the question. Why was it able to conquer all that territory to its east? No other country has done anything even remotely similar. Why could Russia? And how is it possible that it would reach the Pacific Ocean before it reached the Baltic or the Black Seas?!
And why is Russia’s capital not on the Volga, the longest river in Europe? The capital could be Volgograd (“the city on the Volga”), previously known as Stalingrad.
Why isn’t it Kiev, an older city than Moscow, in the middle of the most fertile area in the region, and the capital of Kievan Rus, a predecessor kingdom to Russia today?
Why isn’t it Novgorod or St Petersburg, cities that are much better located for trade, on the Baltic Sea?
Why isn’t it in Novosibirsk, much more centrally located?
How did Moscow evolve from a swampy3 village in the 1100s to the biggest European city with over 20M people?4
An AI rendering of a Medieval swampy town on a river turning into Moscow.
The answer to all these questions is pretty crazy. It involves horse archers, human harvesting, and tiny animals, and it tells us a lot about Russia’s past and its future.
Russia’s capital is very far north. It’s the 3rd coldest capital on Earth,5 and by far the biggest, at 8x the size of the next one.6
It’s so far north it barely has farmland!
Why would you put the capital of the biggest country on Earth on the edge of some of the most fertile farmland on Earth? Capitals need a stable source of food. That usually means farms, or at least a port to import food.7 Moscow is not on a sea port or a major river (we’ll see why later), so it needed farmland to support it. This limits how far north Moscow could be, and it’s pretty much as far north as it can get away with.
Moscow sits in woodlands, and the taiga forest begins north of it. None of this is very conducive to farming. So weird… It’s as if Moscow was trying to escape from fertile southern lands… That’s exactly what happened.
You can see in the maps above that Moscow is north of today’s cropland area, and south of that are grasslands. Why grasslands? Because there’s just not enough rainfall to sustain a forest or crops.
And who lives in grasslands?
Steppe societies are perfectly suited to live in grasslands in a way that no other society is:
Grasslands can’t sustain agriculture without irrigation. In other words, the ground can’t produce enough calories for humans to survive.
But nomads can roam! So they can consume the calories produced across a larger area of grasslands.
This way, they can feed horses, goats, and sheep. They then feed themselves from these animals’ dairy and meat. To give you orders of magnitude, a sheep can feed 100 people for one day, so just 30 sheep are enough meat for a company for a month.8
These animals also gave them wool, leather, furs, and bones for their clothing, weapons, armor, and yurts.
They are extremely mobile. If needed, they could travel up to 100 km per day. Riders had several horses (usually 5-6 mares) and would mount each one in turn to avoid tiring them too much.
This meant they could quickly scout large areas, find weak points, and gather numerous riders there to overwhelm the enemy.
The enemy couldn’t easily retaliate because the nomads would simply withdraw faster than they could chase them.
This is how the Mongols were able to build the biggest empire the world has ever seen in just 70 years.
And how did they treat farmers on these lands? Not very nicely.
The vast expanses of the Eurasian steppe begin about 200 kilometers south of Moscow, stretching from the Carpathians to Mongolia. These black earth soils are perfectly suited for settled agriculture but were mostly uninhabited by peasants up until the mid-17th century, because of the constant threat of nomads’ raids.
The most durable threat came from the Crimean Khanate – a successor of the Golden Horde and the Mongol Empire, and a vassal state of the Ottomans from the late 15th century. The slave trade was one of the main income sources for the Crimean nobility, and an important economic support for the mostly pastoral population. Skilled Crimean horsemen “harvested the steppe”9 by capturing peasants on the Russian, Ukrainian and Polish frontiers, and brought them to the Crimean port of Caffa (modern Feodosia) for export to the slave markets in the Ottoman Empire.
About three million people were captured from all the Slavic lands (Russia, Ukraine, Poland) in the 15th and 16th centuries. About 200,000 people were abducted from Russia in the first half of the 17th century.—All Along the Watchtower: Military Landholders and Serfdom Consolidation in Early Modern Russia, Matranga & Natkhov, 2025. The following quotes are from the same paper.
200k people is about 3% of the Russian population at the time!
With a threat like this, you’d want to attack the enemy to stop their raids… but you can’t. With these types of raids, no farming village could survive. The kingdom of Kievan Rus (the oldest Russian kingdom was based in Kiev10) disappeared because, although it was rich from farming its lands and trading with Byzantium, it was constantly attacked by steppe hordes until the Mongols steamrolled Kiev. Ukraine could have been a superpower, if not for the steppe hordes.11
The steppes were a difficult area for the Russian army to campaign in due to logistical problems. Food could not be acquired from local peasants since they did not exist,12 nor could it be brought by river from populated areas, because the steppes are drained by rivers that flow into the Black Sea, while Central Muscovy is part of the Volga drainage basin, which empties into the Caspian Sea.
Any transportation across the two watersheds would necessarily include slow and expensive portaging across the divide. Furthermore, since Moscow was on the defensive, they had to deploy and feed the guard forces whether nomads attacked or not.
In contrast, the nomadic way of war was perfectly suited to the open and sparsely populated conditions of the frontier. From their winter pastures along the Black Sea coast, raiders would venture north as soon as the spring mud season receded. The scale of the raids could range from a few dozen members of the same extended family to tens of thousands of horsemen. In addition, the raiders could decide when and where to attack, unlike the Russians who had to be constantly prepared for the defense of any part of the frontier. If a location was strongly defended, a band of nomads would seek to distract the Russian army, while others took captives undisturbed.
So it’s not like some king decided to go to Moscow because it was a perfect spot to create a capital. Rather, the alternatives to its north were too cold, and the thriving alternatives to the south were eliminated by steppe hordes.
The band Moscow finds itself in was ideal, because the steppe stops about 200 km south of it, so nomads didn’t have grasslands to feed their animals as they approached. Forests also slow them down, make archery less effective, and provide plenty of wood for protective structures like walls.
This is at the edge of what steppe hordes could reach: Grasses grow with sunlight, so in winter, they don’t grow as much. Nomads had to spend their winters in southern pasturelands close to the Black and Caspian seas.
Travel too close to the winter could be impossible, because the rains create the Rasputitsa (mud season):
Nomads had to raid the north mainly in the summer, and it took about a month to reach the area of Moscow and a month to return, leaving them little time to raid; it was about as far north as they could reach. Moscow didn’t need to defend itself for months on end, it just had to hold off the enemy a few months per year.
Nomads could still reach it, though. Also, the band of southern woodlands stretches for thousands of km east to west. What did Moscow have that was special in that band?
Across the Eurasian Plain, most rivers flow north-south.
It’s inconvenient when your enemies can also flow north-south on horseback because rivers can stop horseback riders, but not if they run parallel to their path!
You need an east-west border formed by rivers. You need…
To its north, the Volga bends westward, forming a strong barrier with its tributaries: The Kama and its tributaries reach the Ural Mountains to the east, and the Oka and its tributaries the Ugra and Moskva reach far westward. And where is Moscow?
Moscow’s historic center is on the north bank of the Moskva! That way, the river protected it from the threat of southern hordes. Moscow built several concentric walls as it expanded outwards, which you can easily see on satellite maps: Like in many other cities, these are now highway rings or boulevards.
Of course, and like in many other cities, this specific point of the Moskva is convenient because the Kremlin could be built on a hill, making it easier to defend:
But this was just the last line of defense. Moscow also had to defend all the surrounding countryside that provided food, wood, and revenue. And this position along the Mokva is ideal because there’s another natural line of defense to its south: the Bereg Line (Riverbank line) along the Ugra and Oka rivers, which cut the invasion axis from West to East. So Moscow constructed a series of fortification lines similar to the Great Wall of China and the Roman Limes from the mid-16th century.
Its strategy was simple: Garrison the Ugra-Oka line so as to block all possible fording locations, repel any river crossing attempts, and wait for the campaign season to draw to a close.
The first time Moscow achieved this was in The Great Stand on the Ugra River, in 1480:
Moscow’s Ivan III was able to garrison every possible ford on the rivers, and stopped the southern nomads. As winter approached and the rivers started freezing, he retired his forces towards Moscow, to ambush the nomads in the forests. This additional buffer was too much, and the nomads turned around in November.
This is a perfect illustration of Moscow’s ideal position: In the wooden range, south enough to get some farming but north enough to delay the steppe nomads, and behind the natural barrier of the Volga – Oka – Ugra –Moskva rivers.
But this is not enough: Moscovites got lucky that year, but nomads could try this every year. They only needed to get lucky once in order to sack Moscow, so in the 1500s, they tried to raid Moscow 20 times! They succeeded a few times, notably in 1571, when Moscow was burned, its population shrunk from 100k to 30k, and nearly 150k people in the region were taken as prisoners for the slave trade!
Moscow needed a deeper defense. So it built it.
If you’re keeping count, that’s three lines of defense at this point.
To avoid the possibility of a single point of failure and to expand the protected area, a new line was built some 50-100 km to the south. The Great Abatis Line (Bol’shaya zasechnaya cherta), also known as the Tula defense line (Tul’skaya cherta), was a chain of fortifications erected until13 the 1560-s about 180 kilometers south of Moscow. The line was built from felled trees laid in a row with the sharpened tops towards the enemy and augmented, where possible, by earth mounds, ditches, and watchtowers. The line was a formidable obstacle since it deprived the nomads of their main tactical advantage – mobility. By 1630, the line consisted of about 40 fort towns and stretched for more than 500 kilometers in the east-west direction. This type of fortification could be quickly built in forest areas. Aside from being an obstacle for nomads and a natural shelter in case of unsuccessful defense, forest was the main source of construction material. This explains why the southern border of Muscovy did not advance past the forest-steppe boundary until the end of the 16th century.
Wherever it could use existing rivers, it did. Elsewhere, it built wooden palisades thanks to the available forest around. Since this area had no farmers due to previous raids, Moscow needed to send soldiers there and feed them for months. That was too expensive, so it granted lands north of it to the soldiers that would defend it. Interestingly, this was the origin of Russia’s serfdom, as described in Matranga and Natkhov’s amazing paper on the topic, which inspired this article.
You can see this on the map of Russia’s expansion:
First, from 1300 onwards, the Duchy of Muscovy expanded northwards. Why?
Sure, it was so cold that nobody else could challenge Moscow in conquering that area. But the fact that it’s too cheap to conquer doesn’t make it worthwhile. Aside from low cost, Russia needed a benefit. What do you think that was? What kind of wealth can you get from a landscape like this?
You get this:
The fur trade was mostly of small squirrel-like animals called sable.
It was controlled by Novgorod, which was connected via rivers to the Baltic.
Unfortunately for Novgorod, it’s close to Moscow, and on the path to the river trade routes between the Baltic and the Black Sea.
Moscow had farmland to grow its population, whereas Novgorod didn’t. The result was that Moscow conquered Novgorod in the 1470s. This is basically the exact same reason why Britain was able to conquer Canada in the 1700s: Canada was dedicated to fur trade so its population was small, while the British Colonies were farmers, so their population was much larger.
Note that, although Moscow could trade with the Baltic, it avoided actually controlling any land on the Baltic Sea itself, because there it was exposed to powerful kingdoms like Denmark, Poland-Lithuania, or Sweden. Instead, Moscow went eastwards.
We’re now well on our way to explain why Moscow reached the Pacific Ocean before the Baltic or Black Seas: more of the same.
Once all the northern hinterland was secured, Moscow had the farming from its region and the income from the northern trade. It was now ready to fund its line of defense to stop the attacks from steppe hordes. But in the 1500s and 1600s, it was not able yet to control the steppes. It would take centuries for Russia to achieve this.
Moscow was also not in a position to challenge much more powerful European countries to its west, including all the Hanseatic League cities. So Russia expanded in the easiest direction left: eastwards. There, it didn’t need to create any new economic model. It has just to expand the one it knew from the north: fur trade.
It took less than one century to conquer everything from the Urals to the Pacific. Why?
We saw it in Russia’s Dilemma: Siberia is too inhospitable.
It’s too cold, too far away from the warming influence of the Gulf Stream
It has taiga forests to the north (and tundra beyond that) and steppe to the south, with only a small stripe that could be used for agriculture.
It’s too far from trade routes, because it has no rivers that flow into trade seas. In Siberia, they all flow into the Arctic Ocean.
No trade means no wealth. No wealth means it was very hard to finance the infrastructure needed to drain the ground and irrigate the fields.
So the region was barely farmed, sparsely populated, and the locals were also predominantly nomads, but much less powerful than those farther south.
Given this poverty, it was easier for Russians to conquer these peoples than the southern hordes.
Add to that the fact that Russians brought diseases with them (like Spaniards to Americans), which killed up to 80% of locals, and you can understand why the Siberian population was not more than 300k people in the 1600s, when Russia conquered them.
So this is why Moscow couldn’t be located farther east:
It controls the Volga River Basin, which is the easternmost river that flows southwards in Eurasia.
East of it, there are the Urals, a huge barrier that made everything east of it poor, as it was disconnected from European markets.
Past the Urals, it was too inhospitable for a large population center to appear.
Why isn’t Moscow farther west, though?
Given everything I’ve told you, it’s obvious that Russia had to be a land-based kingdom. It had to expand across the northern forests, across Siberia, and more importantly, fight the southern hordes. This means all investments went to a land army, so it could afford no strong navy. Hence, Russia had to stay away from the Baltic and Black Seas for the longest times.
It’s telling that, historically, the most Russian city on the Baltic was Novgorod, connected to Baltic trade, yet not exposed enough on the Baltic to get conquered by Baltic powers.
There are also many rivers that flow into these seas, facilitating trade and the kingdoms that emerged from there.
Two examples that illustrate the challenges of powers west of Moscow are the Kievan Rus we discussed before, and Poland-Lithuania, a huge kingdom that marched to Moscow several times.
Technically, Moscow can reach all the way to Germany, where the Northern European Plain shrinks so much that it would be hard for Russia to stretch any further. That means the region between the Oder River and the Volga River was going to be highly contested. At its farthest expanse, the USSR did reach into Germany, but that was untenable given all the advanced societies that emerged in Eastern Europe thanks to its rivers and seas.
To its north, it’s too cold for farming, so no big city could emerge to prevail in the region.
To its south, nomads “harvested” farmers for centuries, emptying the country.
Moscow is in a narrow band between north and south where it’s ideal: South enough that it can farm, north enough that it has woods to protect it and provide wood.
Given the logistics of the hordes, driven by the southern pastures, seasons, and the Rasputitsa, Moscow was just out of reach from steppe hordes, provided that it could defend itself.
For that defense, it also needed rivers, and the only seriously big east-west river in the area is the Volga and its tributaries. Moscow has a double defense there, thanks to the Oka – Ugra and the Moskva.
The Volga reaches into the Urals, protecting from southern hordes there too.
The Volga also reaches very close to the Baltic, so Moscow could expand northwards easily and take control of the Baltic fur trade.
The city that would control the Volga’s water basin would also control Siberia and its furs, because it’s an empty desert that happens to reach all the way to the Pacific.
To the west, you have too much competition from populations that grew around very viable river basins.
Now we can answer the rest of the questions from the intro:
Why aren’t the Baltic cities of St Petersburg or Novgorod the capitals of Russia? Novgorod was richer earlier on thanks to Baltic trade, and St Petersburg is recent but was for some time the capital of Russia. But although rich, both are too exposed to Baltic powers, and are too far north to develop a big local population from farming.
Why isn’t Novosibirsk the capital, much more centrally located? Or any Siberian city, for that matter? Because historically these were very isolated, exposed to hordes, poor, and backwards.
Why isn’t it Kiev? Because it was in the steppes, so too threatened by the hordes.
Why isn’t the capital somewhere else on the Volga river basin? Why on a secondary tributary? Because most of the Volga river Basin is either too far south and exposed to hordes, too far north and too cold, or too far east to interact with European trade and wealth.
Why did it reach the Pacific before the Baltic or the Black Sea? Because Russia had to be a land army, and there was little competition all the way to the Pacific (but a lucrative fur trade), whereas Baltic and Black Seas had massive competition from seaborn kingdoms and empires.
As you can see, the history of Moscow is the history of Russia. Once it became the capital of a transcontinental empire, people from all over the empire moved to the capital to be closer to its power and wealth.
What does all this tell us about the future of Russia?
Moscow is very much embedded in the North European Plain and its system of trade and communications. It pulls it in that direction. Alternative sources of power will always exist closer to Europe, as long as trade flows.
Its granaries are south and west of it though. They are likely to continue having the highest population densities in the country.
Until some sort of trade-based peace is brokered across the Northern European Plain, Moscow’s thousand years of conflicts in the region will make it insecure and fuel conflict.
Siberia is still foreign land. It’s too big, and Moscow is too remote.
But this leaves many questions:
Beijing is much closer to Siberia than Moscow. Will China challenge Russia’s presence in Siberia? When?
Why does Mongolia exist, if nomads have been suppressed by Russia and China after centuries of invasions?
How did trade work in Russia, with so much cold and so few roads for so long?
How could Russia beat nomads so fast in its expansion eastwards in Siberia, whereas it was having trouble near the Black Sea?
Moscow had another big asset: It gained an Orthodox Church seat a full 150 before it could face the nomads or conquer Novgorod. How come?
These are some of the things we’re going to explore in next week’s premium article.
The Moskva ends in the Oka, which is a tributary of the Volga, which feeds the Caspian Sea, which is only called a sea because the Greek called it so, because they didn’t navigate far enough to realize it’s just a lake.
Canada, the second largest, is a full 40% smaller!
The name of Moscow itself comes from roots that likely mean “wet”, “puddle”, “pool”, “immersed”, or “drowned”.
After Kazakhstan’s Astana and Mongolia’s Ulaanbaatar
Stockholm’s metro area has 2.5M people.
Although it’s a risky proposition to rely on imports for the capital, most northern capitals have direct access to the ocean: Finland’s Helsinki, Sweden’s Stockholm, Norway’s Oslo, Iceland’s Reykjavik, Ireland’s Dublin, UK’s London, Scotland’s Edinburgh, Wales’ Cardiff, Estonia’s Tallinn, Latvia’s Riga, Denmark’s Copenhagen… Canada’s Ottawa, which is farther south than Paris, is on the massive St Lawrence River Valley, and has direct access to the ocean. And yet it’s minuscule compared to Moscow. The only exception I can see is Lithuania’s Vilnius, but the country is much smaller than Russia, and Vilnius is farther south than Moscow.
Steppe warriors were subdivided into units of 10, 100, 1,000, and 10,000 — in Mongolian, these units were called arban, jaghun, minquan, and tümen (pl. tümet). Source.
The paper has a bunch of references to support each statement.
Kievan Rus is a term from later Russia.
From Andrea Matranga’s thread on the origin of serfdom in Russia: Say you want to feed 10,000 men. Right after harvest the average farm household will have enough food for about 1800 people for a day (365*5 adult equivalents people). So as long as the army finds 6 new farmsteads each day, the General can keep his men fed. This could mean buying 10% of the harvest from 60 farmsteads, or paying them to give you over everything and move to the city. Or if they weren't your people just loot the whole thing.
But if there's no farmers anywhere, you just can't march there for any length of time. An ox team will eat the whole cartload of oats in just 15 days, so in practice any place more than 150km or so from farmers might as well be on the moon (except for water transport!).
They said “in the 1560s”, but I think they meant until then? Wikipedia suggests that’s the case.
2025-08-12 20:02:55
Each quarter, we apply the UT lens to the world’s chaos to extract clarity. Today, we’re looking at energy, environment, and politics. For premium readers only.
Back to fertility. Here’s a shocking quote:
Low fertility is a false solution to climate change: the population …
2025-08-08 22:03:20
Each quarter, we apply the UT lens to the world’s chaos to extract clarity. Today, we’re looking at:
Fertility rate: Why the drop, and further evidence on how to reverse it
Real estate, especially neighborhood companies
Game theory of relationships: Why women are more into submission than men, the surprising relationship of young women and porn, dad bod attractiveness, gender interest in looks, and needs for romance per sex.
Diversity: How it’s not good for economics, and how the woke pendulum is swinging
The next article will be premium only, and will cover energy, the environment, and politics: Solar vs Nuclear, news on vertical farms, the fall of free speech in Europe, why Europe lags the US in tech, and more.
Japan lost one million Japanese in 2024.
They’ve been countering it with immigration, but it’s not enough.
Although I have hopes that fertility rates improve in the future, they are abysmal right now. This paper has interesting insights. From one of the coauthor’s corresponding Twitter thread:
Fertility is falling everywhere: rich and poor countries alike, booming and stagnating economies, secular and religious societies. The decline is happening far faster than anyone anticipated, even me, ten years ago!
For example, Colombia’s fertility rate is 1.06, Iran’s is 1.44, and Turkey’s is 1.48, all of which are below the U.S.
The decline accelerated around 2014, well before the COVID pandemic.
As a result, humanity’s fertility is likely already below the replacement rate.
Many assume the replacement rate is 2.1 children per woman. That’s true for rich, advanced economies. But not for emerging economies, where selective abortion and higher young female mortality push the replacement rate higher. Thus, for humanity, the replacement rate is closer to 2.2.
Most of the differences in economic growth among advanced economies over the past 35 years can be attributed to demographic factors.
The 2024 UN World Population Prospects are riddled with data and forecasts that, frankly, make little sense to my coauthor Patrick Norrick.
This last point is very telling. The UN has shaved ~300 MILLION people off of its projection of future human population… in just two years!
It’s due to intelligent forecasts like these:
I don’t know who is signing off on these, but if I did this as a consultant, I would have lost my job overnight.
In our last fertility article, I suggest that people are quite rational about having babies: If the pain of having them is lower than the benefit, they will have more children.
Recently, life has gotten more awesome, but childrearing hasn’t, so fertility rates have shrunk.
A new paper reinforces this theory; this is what having children does to women vs men:
Women’s happiness is neutral, their professional work hours go down, and their household work hours increase. Meanwhile, men’s work and chore hours stay the same, while happiness grows. This suggests women are unhappy with the deal they get with childbearing vs men, probably linked to chores.
But in our last fertility article, I also suggested ways that fertility rates might go up through technology. I stumbled upon this article describes the existing research about the Baby Boom and it reinforces this idea. The article concludes there were three main causes to the Baby Boom:
The improvement of household tech reduced chores and freed time for women.
The improvement of medical tech reduced female death rates in pregnancy and giving birth.
The increase in housing supply reduced housing costs.
All these reduced the pain points of having children, and hence increased the return on investment of having them.
This is cause for hope, because it means that if we continue reducing the pain points of having children, people are likely to have more children. This provides massive support to our hypothesis that tech can be a huge driver of fertility. Since tech progresses faster than the fertility drop, I am optimistic.
The three drivers of Baby Boom fertility also suggest how to increase fertility rates in the future. Since we don’t want to reduce quality of life, we’re left with reducing the cost of having children, as happened back then:
As proposed, improve fertility tech in-vitro gametogenesis, foetus screening, artificial wombs, etc. That is the equivalent to the Baby Boom’s improvement of medical technology.
As proposed, improve tech to make parenting easier: robots and AI for nightcare, childcare, education… This is the equivalent of the Baby Boom’s automation of chores.
Housing! If, like in the Baby Boom, housing costs shrink, fertility will increase. As you know, I think this might be a natural consequence of a shrinking population. But ideally we wouldn’t wait decades for that to happen. So if we want to increase the population, we should make housing cheap, which means massively increasing the supply of new houses!
In Why I Don’t Invest in Real Estate, I shared that the main drivers of increased demand for housing were urbanization, the increase in population, the decrease in family size, and the size of our homes. I think this picture captures this last point perfectly:
This is Manhattan in 1931, a city with very few skyscrapers, mostly 3-5 story buildings, and more population than today! That’s what our shrinking family sizes and bigger homes have done.
In The Fundamental Problem with Urbanism, I shared the idea that urbanism might be improved if big companies own full neighborhoods. One way to do this is through transit-oriented development, like in Tokyo1, one of the few modern cities in the world that is dense, populous, and livable. There, companies and the public sector buy land and build both the transit, the station, and the land around it. In this setup, companies are incentivized to make that land as valuable as possible, so they pack it with amenities.
Women are much more interested in being dominated than men are interested in dominating:
This dominance gap is weird, isn’t it? I can imagine why men are aroused by it: Evolutionarily, a violent man could force his way into sex more than a non-violent man, therefore reproducing more. I can also imagine why women are aroused by it, for the same reason: They would sire sons that are more likely to have children.2
The conundrum is why would women have this fantasy more than men? After all, forcing into sex is likely to have more downsides for women (having children of fathers they don’t like, being hurt, dying) than men.
2025-08-07 01:12:11
Catching up on articles from the previous weeks, today we’re covering the most important things that have been happening in the world of Robotaxis and AI since we last discussed them six months ago.
In Robotaxis Are Here, I explained why they’re arriving, why they’ll broaden their spread in the next 1-3 years, and why Tesla is best positioned to win that market. As a fast-moving trend, we need to track it closely.
Here are Waymo’s weekly rides, just in California:
And adoption is faster in Austin:
That makes sense: A lot of hesitance comes from ignorance. The more Waymo proves itself and people learn about it, the faster they’ll adopt it. Soon enough, we’ll see people begging Waymo to enter their market.
Waymo just entered Atlanta, by the way, will open soon in Miami and DC, and is collecting data in Las Vegas, Dallas, San Antonio, Nashville, New Orleans, and several other metros. It now takes only 1-2 years from mapping vans to credit-card rides.
And Waymo is destroying the competition. It has surpassed Lyft in rides in SF, and is on track to surpass Uber within 8 months or so:
And this is with Waymo taking 2x longer and costing 70% more than Lyft!!!1 That’s how much better the Waymo experience is: People really care about not having a driver!
But the bigger story is not just that they’re going to crush the competition. It’s that robotaxis will make the market size explode. Uber said ride-hailing could grow by 25x if its price dropped under $1/mile.
By replacing other private rides with ride-hailing:
Uber couldn’t make it happen. But in Austin, now Tesla costs $1 per mile:
As a comparison, ride hail customers are currently paying nearly $3/mile…2
If Tesla maintains this type of pricing, it won’t make sense for drivers to continue their job, and Uber and Lyft will crash.
8% of US workers are professional drivers.
I maintain that robotaxis will first grow by replacing cabs, then by taking new use cases that were previously impossible—like the one below. I’m not sure whether it’s a joke or not… Yet. If this tech works reliably, it’s a matter of time before it becomes more widespread:
A key input is going to be the cost of each car: The cheaper they are, the cheaper the price tag can be, and the more competitive they will be. And Tesla’s cars will be nearly an order of magnitude cheaper than Waymo’s:
According to this Tesla analyst:
In the recent testing of 36 vehicles in China, vision-only Full Self-Driving beats all the leading systems with LiDARs incorporated. Separately, CEO Robin Li recently alluded to Baidu’s transition to vision-only approach while Xpeng is in the early stage to phase out Lidars.
Are politicians criminals for slowing down the adoption of robotaxis, when these cabs reduce crashes and injuries by one order of magnitude?
In clinical trials, when it's obvious that a treatment is much better than the tested alternative, it's considered inhumane to continue subjecting people to that alternative. There are rules for early stoppage of clinical trials. Shouldn't we do the same here?
With the data we have today, we should not be saying "Let's be cautious with robotaxi rollout. We should be asking: "What can we do to accelerate their deployment? Millions of lives are at stake."
I didn’t realize how important this is until I read this article:
Something like 40,000 people die in traffic accidents in the US every year. The number is over one million per year globally.
There are over 5 million non-fatal injuries from car crashes each year that require medical attention in the US.
In 2010, the total costs from these events was $836 billion, or ~$2700 per American per year.
But these costs are just the tip of the iceberg because most of the cost of transportation, at >$2 trillion per year, comes from adjusting to human inadequacies.
Wait, what? Car accidents are costing trillions to the world economy?3 How?
A big share of the materials in cars are due to safety. Without accidents, you can strip them out, saving all their money. Austin Vernon calculates we could make car weights 10x lower.
Automobile shapes today trade off safety and aerodynamicity. Without safety, they can become more aerodynamic, and move faster at a cheaper cost.
Cheaper transportation costs massively improve the economy.
Lower weights on roads means less road wear, and hence less maintenance cost.
Many from the same article:
Robotaxis are much more likely to be electric, because that will make them cheaper per km. This means much less oil consumption, and oil countries will suffer accordingly.
Goldman Sachs says insurance costs will be halved. It’s probably much more than that, because cars that have fewer accidents will also be lighter, so the gravity of their impacts, when they happen, will be lower.
Looking for parking will be a thing of the past. That’s a $150B industry worldwide. This also means homes and apartment buildings will be cheaper: Building a parking spot costs $28k. And streets won’t be littered with parked cars anymore. We will be able to reclaim this space for pedestrians. On the flip side, cities will lose a huge source of income. NYC currently makes nearly $1B/year from parking meters and parking violations!
Maintenance of electric cars is already much lower than for internal combustion engine cars. If you add to that lower car costs, demand for them will increase. The economies of scale will mean cars will be so cheap, it might make more sense to replace them than repair them. Mechanics basically disappear.
Pollution plummets: not just combustion, also tires and road wear, which account for a big part of pollution!
Noise plummets. City streets get reclaimed, with more people walking on them and sitting on the sidewalk to hang out.
People will own several of these: for groceries, children, food takeaway…
Poor people will have an access to transportation previously unheard of.
It might be easier to have more children, as children’s extracurriculars won’t entail parents acting as cab drivers all day long.
Public transportation might become outdated.
Some cars will be adapted to commutes (more aerodynamic), others to older people (more ergonomic), others to other groups.
Learning to drive will be a thing of the past.
Traffic jams will shrink, as this type of car will be smaller, and AI drivers will need less safety distance.
Trucks lose the cabin, and become just a surface with wheels
All in all, it feels like Waymo is trying to expand fast in the US, it will gain tremendous market share as it does, but it won’t fundamentally undermine the model of robotaxis. If Tesla succeeds in its markets, it will drop the price, drive competitors out, take over the market, and expand it by at least one order of magnitude. As this revolution takes place, our daily lives will change dramatically.
OK, now let’s look at AI: Where it is, where it’s going, the risks, and more.
2025-08-01 02:34:45
Each quarter, we apply the UT lens to the world’s chaos to extract clarity. Today, we’re looking at all the interesting GeoHistory themes that have emerged recently. This is not meant to be read cover to cover, but rather for you to jump to the topics you’re most interested in. Today:
The Economics of Space
Starbase
Trade Wars
Phantom Maps
Druzes vs Syria
Forgotten civilizations in the Sahara?
A humorous and revealing map of Brazil
The volcanic fall of the Roman Empire
Poland vs Russia
English as the Lingua Franca of the world
Mountains vs flatlands
Why do we see what we see, and not other types of light?
An 11 meter painting that reveals a lot about China’s past
Influencer strategy, 120 years ago
We were right about global taxes
In No Room for Deep Space, we saw that the economics of space mining are quite bad, because there aren’t that many rare minerals easily accessible in space, and space mining is too expensive. If there were a very valuable element to mine in space, its quantities would need to be such that it would flood the market, crashing prices, and making the entire venture senseless. We saw a good example of this recently.
As far as I can tell, the original source of this claim is this Forbes article, which quotes no source or calculation. My interpretation is that scientists claim this is mostly a metallic asteroid with 10 quadrillion tons of mass. To give you orders of magnitude, that’s about 10 million years of iron production on Earth.1
Journalists speculate that if we brought this asteroid back to Earth, the quantity of metal available would be of gargantuan value: Iron is worth $424 per ton, nickel $14k, and gold $75M, so as long as it’s mostly metal, the asteroid would be worth many quintillion dollars, regardless of its composition.
At current metal prices.
But of course, these metal prices would crash if we had such an asteroid available to mine. In fact, if it were indeed very high grade ore, virtually all metals it contains would become nearly worthless overnight.
That said, this makes me think this could actually be a valuable project for the world, if we figured out a cheap way to get this metal to Earth. Yes, those metals would become worthless, making the venture a money-losing proposition, and killing many mining industries in the process. But what would happen next?
Nearly all products that use these elements would suddenly drop in price. Across the world, many expensive things would become cheap, as their cost converges towards that of energy and logistics. Demand for these cheaper products would skyrocket. It would usher in a new era of abundance. What would that look like?
Steel would become even more ubiquitous. Housing would contain much more steel. Big infrastructure like bridges, electric grids, ships, robots, rockets, and airplanes would drop in price. Mining would shrink, and with it, its ecological footprint.
Since we’re talking about space: On January 16th of this year, we said Boca Chica should become a city. It became one in May!2
In What Asian Development Can Teach the World, we discussed how Japan, Taiwan, and South Korea navigated the world economy after 1945 to become rich countries. There were three key strategies:
Redistribute land so that farmers are owners, care more about the output, and farming yields increase.
Support and protect core private industries, under the condition that they must win in internationally competitive markets. Kill the companies that can’t win internationally.
Intervene in financial markets so that people’s savings are used to finance point #2. Liberalize them little by little, once the industrial giants are well established and the economy can compete globally. Avoid too much debt and too much consumption.
Very clearly, China has been following this path. This is why Chinese people can’t invest abroad, why the stock market there is stifled, and why the government let the real estate bubble collapse, while at the same time propping up industries like drones, batteries, and solar panels. Internal consumption used to be discouraged.
In this interview, Kenneth Rogoff (the former Chief Economist of the International Monetary Fund) highlights what happens when you have 1 and 2, but too quickly move to #3. This happened to Japan when it signed the Plaza Accords in 1985. Seven years later, Japan’s real estate and stock market bubbles burst, the country plunged into a generational crisis, and Japan has not recovered since.
According to Rogoff, Japan could be 50% richer than it is today if it hadn’t followed that path. A country that was richer than the US will soon be poorer than Poland.
What happened? The Plaza Accords came from US pressure on Japan to strengthen the Yen quickly. Otherwise, Japan was exporting too much; it was too competitive. The US also pressured Japan to liberalize its financial markets.
The yen doubled in value in three years. This caused a trade reversal, with exports crashing and imports spiking, and lots of companies suffering heavily. To offset the economic shock, Japan slashed interest rates and flooded the economy with cheap credit.
This meant that Japanese banks had plenty of cash and a liberalized market; they went on a lending spree. They poured money into real estate and stocks with little risk assessment. Japan's stock market became worth more than the US stock market despite having half the population. The total value of Japanese real estate was 4x that of all US real estate!
When the bubble burst in 1991, banks were left with massive amounts of bad loans. The entire financial system crashed, creating a "lost decade" of deflation and stagnation, and Japan has yet to fully recover.
This is helpful context for understanding the trade war between the US and China, and why China doesn’t want to increase the value of its currency. That’s probably one of the reasons why the US prints a lot of money, in a process that depresses the USD’s relative value and boosts its exports.
And this is not just a fertility or aging issue; Japanese workers are just less productive:
Even if you only look at productivity per hour worked:
One of the tools Japan, Taiwan, and South Korea all used to protect their industries was tariffs. But this is easier said than done: Tariffs can cause a lot of harm. Did these countries’ tariff approaches work? Or did their economies grow despite them?
These countries prioritized infant industry protection, to protect young but important industries. Their governments had used the example of Germany and the US, which had used tariffs just as they had heavily industrialized. But according to this paper, tariffs didn’t help. Instead, they generally:
Reduced labor productivity
Made more, smaller companies
Raised prices
Employed more people
Created more value internally
How? By weakening foreign competition, more local companies could exist, create value, and employ people, but that’s because companies that weren’t very productive could survive. This lack of productivity was reflected in higher prices.
My shallow take on all of this is that protectionism for defense and infant industries is still valuable, but should be avoided for others. It also means that it makes sense to tax downstream products rather than upstream:
Don’t tax things like ore, steel, cement, or heavy machinery, unless you want your country to develop it (whether for geopolitical or economic reasons).
The closer a product is to end consumers, the more sense it makes to levy tariffs:
They hinder consumption (so people do more saving, which can be reinvested in growth)
They don’t hurt other industries downstream
We’ve done phantom maps of Poland and Germany. I stumbled upon this phantom map of ex-Yugoslavia, showing how illiteracy strongly correlated with Ottoman control:
Did mountains influence Ottoman presence, which in turn determined literacy?
Next:
Druzes vs Syria
Forgotten civilizations in the Sahara?
A humorous and revealing map of Brazil
The volcanic fall of the Roman Empire
Poland vs Russia
English as the Lingua Franca of the world
Mountains vs flatlands
Why do we see what we see, and not other types of light?
An 11 meter painting that reveals a lot about China’s past
Influencer strategy, 120 years ago
We were right about global taxes
2025-07-28 18:12:08
I’ve been a bit delayed on articles the last couple of weeks. I will catch up in the coming days.
What’s your staple, bread or rice?
This is a momentous fact, for it might have determined politics, culture, and wealth.
How? Well, bread comes from wheat, and rice from… rice. Here’s where they’re farmed:
Wheat and rice are not harvested in the same places. Rice and bread are the predominant food where rice and wheat are respectively the predominant crops. Here’s another way to look at the same data:
This, in turn, is determined mainly by this:
Rice crops vs rainfall, side by side:
But this doesn’t fully explain it since it also rains a lot in Ireland, for example, but nobody grows rice there. You need the heat found closer to the equator: Rice grows in hot, wet, flat, floodable areas, whereas wheat prefers cooler, drier, better drained areas.
Flooding rots wheat but can 3x the yields of rice.1 That makes wheat well adapted to hills, whereas rice can only survive on hills when they are terraced:
This sounds like just a fun fact, but it ain’t. Because rice generates twice as many calories per unit of area.
This means that rice nourishes families on half the land that wheat requires.
Which means population density in rice areas can be twice as high as in wheat areas, or four times with double cropping.2 A hectare of land can feed 1.5 families with wheat and 6 with rice.
Yet rice paddies also require a lot of work—twice as much as wheat. And that work is almost year-round: preparing paddies, raising seedlings in nurseries, transplanting every single seedling by hand into flooded fields, managing water, pumping it,3 weeding,4 harvesting, and threshing—often followed by a second rice crop or a winter crop. These tasks peak during transplanting and harvest, creating critical seasons where a huge amount of work must be done in a short window of time.
Crucially, this labor cannot be delayed—if you miss the planting window or harvest late, the crop is ruined. As a result, rice farmers developed reciprocal labor exchange: neighbors help each other transplant and harvest in time. The timeliness pressure meant rice villages became tightly cooperative communities to ensure everyone’s fields were tended before it was too late.
"If one is short of labor, it is best to grow wheat" —Shenshi Nongshu (Master Shen's Book on Agriculture, 沈氏农书), 1600s.
Wheat farming historically had a more seasonal rhythm with periods of relative quiet. Wheat is typically sown in the fall or spring and then mainly just left to grow with the rain. Aside from episodic weeding or guarding the fields, there was less continuous labor until harvest time. Harvest itself was a crunch period requiring many hands with sickles—European villages would collaborate during harvest, and farmers might hire extra reapers.
These differences made these regions diverge across politics, culture, and economy.
The fact that wheat can grow just with rain means less investment in irrigation infrastructure, so individual families could fully handle their own fields. Decentralized states were the result of that, very obvious during Europe’s Middle Ages. That was not the case for rice, which required controlled irrigation and flooding, which means irrigation canals, dikes, reservoirs… all of which require collective effort. Neighbors upstream and downstream had to coordinate water usage; entire villages synchronized planting and flooding schedules. And let’s not forget the terracing. This can more easily give rise to hydraulic societies, highly centralized states that control and coordinate the irrigation system, as we saw in How Rivers Shaped States.5
Maybe this made cultures more or less individualistic? Westerners are famously more individualistic than East Asians. Christianity emphasizes the value of every life and every soul, which leads to more individualistic societies that guarantee individual rights of property and life. Meanwhile, Confucianism, developed in China, emphasizes social duties, puts the interests of society at the forefront, and created societies that are much more sensitive to losing face. Could these differences be linked to crops? When your life depends on your relationship with your neighbors, you better develop a culture of group harmony, family loyalty, and consensus decision-making.
This is the Rice Theory of Culture. But is it true?
To test it, academics went to China,6 a country that is politically, religiously, linguistically, and ethnically uniform, but where the north has historically farmed more wheat and the south more rice. There, they found that:
In psychological tests, people coming from rice-farmed areas (“rice people”) were more culturally interdependent than wheat people (e.g., rice people self-inflate less).
Wheat people reason by focusing on individual elements, while rice people think more holistically.
Rice regions have tighter social norms than wheat regions.7
This comes with stronger social order, less crime, and less drug abuse in rice regions.
But also with less individual freedoms and less acceptance of immigrants. Rice regions are more likely to support authoritarian governments.
Outside of China:
Rice-farming nations had tighter social norms than wheat or herding nations.
People in rice-farming villages in Japan were more concerned about their social reputation than people in fishing villages.
Rice-farming societies tend to have less flexible, less mobile relationships.
Wheat people rate family as less important in life than rice people.
Wheat people are also more likely to think that parents have to earn respect from children, rather than respect being automatic.
Rice people treat their friends better but strangers worse than wheat people.
All this could just be correlation, not causation. Luckily, this paper found a quasi-random test, courtesy of the Chinese Communist government, who quasi-randomly assigned people to farm rice or wheat in two state farms that are otherwise nearly identical.
The rice farmers show less individualism, more loyalty/nepotism toward a friend over a stranger, and more relational thought style.—People quasi-randomly assigned to farm rice are more collectivistic than people assigned to farm wheat, Talhelm & Dong
Outside of harvest and planting, wheat farmers had more off-season time. This free time could be spent on other pursuits: tending livestock (common in wheat regions, since dryland grain and pasture were complementary), crafting tools or goods, or engaging in trade and markets during the winter off-season.
More importantly, wheat areas might have accelerated the Industrial Revolution and influenced the wealth of countries today.
When the Industrial Revolution really picked up, in the early 1800s, most available land in the Old World was already farmed. But not in the New World. Crucially, wheat doesn’t require many workers, so expanding farming in the US, Argentina, and Australia was extremely fast.
Since there was little land limitation, labor limitation really hurt, and that’s one of the reasons why American innovation massively contributed to farm labor automation: Automating the little work required could unlock thousands of square miles of new fields.
None of this was possible in the Old Worlds of China or Japan. The areas that could be opened up to rice farming, like in Thailand when new canals opened up the heartland, did get rice farming. But there, the requirement of labor was so massive that it was quite slow. Many Chinese farmers moved there, but that was not enough. The amount of automation needed to unlock lots of rice farming was too high. It was also much harder work to automate, as machines don’t easily deal with mud. All this meant that mechanization of rice farming came much later, that rice production didn’t increase that fast early on, that when it did it couldn’t easily fuel a new class of rich farmers, and that it did not locally accelerate the need for automation early on.
Wheat grows in drier, colder areas than rice and requires much less labor, but also produces less calories per unit of land than rice. As a result, rice areas had:
More population density
Stronger centralized states
A psychology and cultures that foster social harmony and collaboration
Meanwhile, wheat encouraged the colonization of the New World, allowed it to grow its wealth through farming fast, and accelerated the development of the Industrial Revolution, which increased the economic divergence between wheat and rice areas.
In other words, climate determined crops, which then heavily influenced our societies. Even decades after most of us have stopped farming, these effects carry into our subconscious cultures.
Does this mean these crops fully determined the history of the world? No. But they nudged it in a particular direction, like dozens of other factors that we explore in Uncharted Territories. The world is made of systems that mold us in certain ways, unbeknown to us. It’s only when we realize these influences and systems that we can reclaim them and decide where we want to take humanity.
Via this article: Dryland rice produces 1.2 tons per hectare, but flooded rice produces 5 tons (Khush, 1997). And of course, flooding eliminates weeds.
Double cropping was possible after the introduction of the Champa Rice variety, during China’s Song era, around 1000 AD. I am now connecting the dots and realizing that the population explosion that happened in China around that time, which we discussed in this China article, was at least in part caused by the introduction of this variant, as double cropping would allow for twice the amount of grain harvested! That also made the Song possible (as they only controlled the southern part of China). This population boom also caused the need for currency, which the Song solved by inventing paper money!
This great article mentions that pumping water was sometimes done by pedaling, and could take as much as 70 hours per worker. Fun fact, I once spent a summer in Burma trying to sell water pumps to farmers. We ended up building a microfinance institution.
Flooding reduces weeds, but doesn’t eliminate them all. And weeding underwater is much harder to do.
These societies were more common in high density areas like China than in low-density ones like Thailand. Also, as we saw in the article linked, it’s not the case that irrigation meant hydraulic societies. Egypt was one, whereas upstream Mesopotamia was not, because the ability to predict harvests in Egypt made taxation easy. I assume in the flat and heavy-rain areas of the Ganges, Yangtze, Yellow, Red, Irrawaddy, Mekong, and Chao Phraya river valleys, production was easy to predict, so taxation could be effectively executed, and it would lead more consistently to hydraulic societies than in the less predictable Mesopotamia. But I don’t know, so take this with a grain of salt.
This Reddit comment posits that millet was likely a more prevalent crop in China in antiquity. I don’t think it changes much, because from what I can gather, millet’s requirements and growing regions are similar to wheat’s, and northern China is mostly a wheat and bread region, while southern China is the rice country.
This is not the only, or even main determinant of tightness. Others like external threats are stronger.
