1 November, 23:31
I am convinced that the church and the capital follow the caravan. Trade arteries come first, and centers of power either shift onto them or, conversely, bend those arteries to themselves.
Why count how many people live around a city's center at all? Because administrative borders lie. Moscow spent years annexing land all the way to the Kaluga region line. Chinese "cities" include rural counties the size of a European country. Paris, inside its ring road, looks like a provincial town. Comparing such figures is like comparing harvests measured in sacks of different sizes. So Alessandro Palombo, the author of the chart, picked up an ordinary compass. From the city hall or main square of each city he drew a circle with a 30 km radius, or 2,830 square kilometers, and counted the people inside. The data come from GHS-POP 2025, the population grid of the European Commission's Joint Research Centre, which splits the planet into 1 km squares. Around 220 large cities were compared, and to avoid counting one metro area twice, a simple rule applied: if a city's center falls inside a bigger neighbor's circle, the two are merged.
But the method is not the most interesting part of this study. The question it raises is. What exactly pulls 15, 20 or 30 million people into one circle? A market that gives them work? Or a power that drags resources in from a vast territory and feeds whoever sits close by? The answer tells you what a city really lives on: what it makes and sells, or what it takes.
Thirty anthills and the blank spaces on the map
Delhi tops the list with 30.9 million people inside the circle. Then come Shanghai with 27 million, Dhaka with 25.7 million, Jakarta with 25.6 million and Cairo with 25.3 million. The next five are Manila (21.6 million), Karachi (21.3 million), Tokyo (21 million), Kolkata (20.4 million) and Seoul (19.4 million). After them: São Paulo at 19.1 million, Beijing at 18.2 million, Bangkok at 17.7 million, Mexico City at 17.2 million, Shenzhen at 16.5 million, Kinshasa at 16.3 million, Guangzhou at 16.2 million and Bangalore at 16.1 million.
Moscow is 19th, with 15.3 million. Behind it come Ho Chi Minh City (15 million), Mumbai (14.8 million), Lahore (14.5 million) and Istanbul (13.4 million). Lagos and New York tie at 12.3 million each, Lima and Luanda at 11.5 million each, Chennai has 11.4 million, and London and Buenos Aires close out the thirty at 11.1 million apiece.
Sort the list by region and the picture sharpens. East and Southeast Asia supply 10 cities, South Asia 8, Latin America and Africa 4 each. Russia and Turkey supply 2. The Western world also supplies just 2: New York and London.
Hence a tempting formula: the anthill is a mark of the Second and Third Worlds. Rich countries don't herd people into giant circles; poor ones do. But three cities of the Euro-American set fall out of the formula: Moscow, New York and London. One caveat matters. Moscow's 15.3 million is not the population of its metro area. It is the number of people inside the 30 km circle. The circles are the only honest thing to compare.
Rich Asia breaks the formula too, in the shape of Tokyo and Seoul. Yet both were born as places of control, not markets. Edo, the future Tokyo, became a city of a million in the 18th century for a simple reason. The shogunate forced every feudal lord, or daimyo, to live in the capital every other year and to keep his family there permanently, as hostages. Provincial nobility spent much of its income in Edo. In other words, the city was fed not by trade but by forcibly hauling the nobility and its money to the center. Seoul, from 1394, was the capital of the Joseon dynasty, which lived off a tax on rice paddies.
The chart is really interesting, though, not for who made it in, but for who didn't. No Rome, the first European city to hold a million people. No Paris. Not a single German city. No Spain, which shipped a whole continent's silver for two centuries. No Vienna, one of the five largest cities on earth in 1910. No Kyiv, and nothing at all from Eastern Europe between the Baltic and the Black Sea. All of Africa, with more than 1.4 billion people, yields only 4 cities. And the Middle East, which caravans crossed for millennia, yields only Cairo and Istanbul.
So the question needs widening. Not "why Moscow?", but why some lands gather people into anthills while others, even rich trading ones, do not.
The boring explanations first
Before reaching for deep causes, the simple ones deserve an honest check. There are at least four.
The first is the population base. A megacity is impossible without an enormous country behind it. India has more than 1.4 billion people, China about the same, Indonesia around 280 million, Pakistan around 250 million, Bangladesh more than 170 million. That explains the 5 Indian cities and 4 Chinese ones in the top 30. France has 68 million people, Italy fewer than 60 million, Austria 9 million. There is simply nowhere to draw a 20-million circle from on a base like that.
The second is timing. South Asia and Africa are living through the great exodus from the countryside right now, before our eyes. Europe went through it in the 19th century and settled down long ago.
The third is density. A rich city spreads out. Suburbs of detached houses and cars push people beyond the 30 km line, so a rich megacity always looks smaller inside such a circle than it is. Los Angeles is smeared across dozens of kilometers. Dhaka is compressed.
The fourth, and the most important, is centralization. In some countries one city takes everything: Bangkok, Manila, Dhaka, Cairo, Lima, Buenos Aires, Seoul, Tokyo, Moscow, London. In others, the population is spread across dozens of centers. Germany's Rhine-Ruhr region holds around 10 million people, but they are scattered across a dozen cities along the river, so no single circle makes the top 30.
This is where the simple explanations run out. Because centralization is neither geography nor demography. It answers the question of who collects resources, and from where. If one point pulled in taxes from a vast territory for centuries, it pulls in people today. If there were many such points, or the collector sat somewhere else, there will be no anthill.
Port, river and the continental interior
The idea of sorting giant cities into three categories came from the Russian-language Telegram channel "Dengi i Pesets", and it set the frame for this discussion: an ocean port, a port on a large navigable river, and a big city deep inside a continent.
The first two are easy to explain. Big water means cheap transport, which means food can be shipped in from far away when the harvest fails. Shanghai, Jakarta, Manila, Karachi, Tokyo, Mumbai, Chennai, Lagos, Luanda, New York, Istanbul and Buenos Aires sit on the sea or in estuaries. Cairo on the Nile, Kolkata on the Hooghly, Dhaka in its delta, Kinshasa on the Congo, Bangkok on the Chao Phraya, London on the Thames: these are rivers ships can sail.
But how did Delhi, Beijing, Moscow, Mexico City, São Paulo, Bangalore and Lahore make the list? The first question here is very simple: what did these cities eat, and what do they eat now?
At this point I think a second sorting is essential, by function rather than geography. A city deep inland can be an agrarian capital standing in the middle of its own grain. Or it can be a point to which grain is hauled by force.
Lahore is the first type. It sits at the heart of the Punjab, the breadbasket of the subcontinent, on one of the most fertile plains on earth. So it naturally became an administrative center, then the capital of the Mughals and of the Sikh state. Its elite sat on its own grain instead of hauling it in from a thousand kilometers away.
Mexico City is an agrarian capital of a special, engineered kind. There were no fertile plains, only a swampy basin full of lakes. The Aztecs solved the problem technically: they built chinampas, artificial "floating gardens" of lake silt and reeds. Yields were enormous, enough to feed Tenochtitlan, a huge city for its time with around 200,000 people. The Spaniards who arrived in the 16th century turned the Aztec capital into the hub of their colonial bureaucracy. And here the function changed. The agrarian center became the point where the viceroyalty's silver and taxes were collected, with roads from the ports of Veracruz and Acapulco converging on it like spokes on an axle.
The largest cities of inner Eurasia, however, did not grow because trading or farming was easy there. All of them lay in zones of risky agriculture. They grew because tribute was easy to collect there. They are textbook fiscal and administrative centers of conquerors.
Beijing was the capital of the Jurchens, the Mongols and the Manchus, all of whom came off the steppe. For them it was the ideal point from which to run settled China: close to the home steppe, yet at the center of the resource base. The local land never produced enough grain for a city that size. Food came up from the south along the Grand Canal; without it, a city of a million would simply not have survived. Note a detail that will matter later: it was the Mongol Yuan dynasty, at the end of the 13th century, that ran the canal straight to Beijing. The capital did not move onto a trade route. The route was dug to the capital.
Delhi was the capital of the sultanates and of the Mughals, all of Turco-Mongol origin. Same logic: an incoming elite settles on the Indo-Gangetic plain and builds a machine for extracting taxes from peasants. The Grand Trunk Road, which Sher Shah Suri restored in the 16th century from Bengal to Kabul, ran through Delhi and Lahore for a reason. It was a collector's road, not a merchant's.
Moscow grew on the right to collect tribute for the Sarai Horde, the Golden Horde of the steppe. When the Horde weakened, Moscow appropriated its fiscal machine and kept collecting, now for itself. It even took the Horde's title. In 1547 Ivan IV assumed the title of "czar", the word the chronicles used for Batu Khan and the Crimean overlords, while his grandfather Ivan III had still been styled a grand prince.
Tehran fits this logic too, although it did not make the top 30. It became the capital at the end of the 18th century under the Turkic Qajar dynasty, as a center for controlling the settled population and the trade routes at the foot of the mountains.
So in all these cases the city is not a "market". It is the point where an elite concentrates resources.
Two special cases remain. Bangalore grew out of a British military base. In 1809 the British set up a cantonment here to control southern India. Railways and engineering schools followed. After Indian independence, factories were placed on that frame, and at the end of the 20th century English-speaking engineers unexpectedly became the city's main export. A 19th-century garrison accidentally laid the foundation of a 21st-century IT hub.
Finally, São Paulo is the only city in this group that grew more or less along market lines. With a caveat. In the 19th century it became the center of a coffee boom. Railways, a huge flow of immigrants and the port of Santos 80 km away made it an industrial hub. But this "market miracle" rested on slave labor. Until 1888, São Paulo's coffee plantations ran on enslaved Africans. Once slavery was abolished, the planters switched to mass imports of European immigrants. São Paulo is thus a blend of an export commodity, coffee, and exploitation, in which a change in the legal status of labor did not change the substance: the city was built on the cheapest, most rightless labor available. That is what let it become a "giant in the continental interior", though only nominally, since 80 km from the sea is not very far.
The channel's conclusion: a continental megacity is almost always a product not of trade but of power, or, in rare cases, of a very particular economy.
Ports as pumps
I would add the ports to the picture. A large share of the coastal anthills on this list did not grow "by themselves" either. Foreign hearts of capital built them as pumps for extracting resources. Batavia, today's Jakarta, was founded by the Dutch East India Company in 1619. Bombay, Madras and Calcutta, the Mumbai, Chennai and Kolkata of the chart, were built or expanded by the English East India Company in the 17th century. In 1571 the Spaniards made Manila the terminus of the silver galleon from Acapulco, which for more than two centuries carried American silver to China. From 1576 Luanda was Portugal's port for shipping slaves to Brazil. Shanghai became a treaty port after Britain's Opium War victory in 1842, Britain annexed Lagos in 1861, and France seized Saigon, today's Ho Chi Minh City, in 1859. Lima was the capital of a Spanish viceroyalty, and Potosí's silver passed through its port, Callao.
Picture the chart as an X-ray. It shows more than big cities. It shows the old pipes of past hegemons: wherever a pump once stood, an anthill stands today.
The port-pump and the continental collector, in other words, work on the same principle. Both pull into a single point what is produced across a vast territory. The only difference is direction: the collector keeps what it gathers, while the pump pushes it onward, across the sea. That is why an anthill appears in both cases. Around any point through which other people's labor flows, the people who service that flow always gather.
The church (a bank) follows the caravan
Now the most important part. I am convinced that the church and the capital follow the caravan. Trade arteries come first, and centers of power either shift onto them or, conversely, bend those arteries to themselves.
To see why the church senses so precisely where the money will turn, recall what it was. The church is a bank. A better one than any bank, in fact. It collected the tithe, a standing tax on every household. It took deposits "for the repose of the soul" that nobody could ever withdraw. It lent to princes and kings. And above all it had what no lay banker had: immunity from taxation and from confiscation. The banking houses of the Bardi and the Peruzzi went bust in the 1340s, the Medici bank closed in 1494, and the church never went bankrupt once.
The Medici books show best how this bank worked. Their Rome branch, which served the papal curia, produced up to 62% of the whole bank's profit. In 1427 that branch held around 100,000 florins in papal deposits, while the entire capital of the Medici bank was only about 25,000 florins. The biggest bank in 15th-century Europe was, in effect, back-office staff for the church's money.
And a bank always goes where the money flows. Let's test this against the history I know best.
While the main artery of Eastern Europe was the river road "from the Varangians to the Greeks", running from Scandinavia to Byzantium, the center sat on the Dnipro, in Kyiv. But in the 12th century the artery began to dry up. The Crusades opened a direct sea route to the Levant for the Italians, and trade with Byzantium moved across the Mediterranean, bypassing the Dnipro. Tellingly, in 1169 the troops of Andrei Bogolyubsky, prince of the Vladimir-Suzdal land, sacked Kyiv, yet the prince himself did not take the Kyiv throne. He stayed in Vladimir on the Klyazma, in the Volga basin, where a different trade ran through Volga Bulgaria to the Caspian.
After the Mongol invasion the Volga became more than a trade route. It became the route of tribute. On the Volga stood Sarai, capital of the Golden Horde, where the levy from every conquered land was delivered. From 1243 the princes traveled there for yarlyks, the khan's patents granting the right to rule and to collect. And here is what the church did. In 1267 the Horde's czar Mengu-Timur issued the metropolitan a yarlyk exempting the church from tribute. In return, it was to pray for its czar in Sarai. The church became the only institution sitting inside the Horde's tax machine on privileged terms. The bank, in other words, obtained offshore status inside the conqueror's tax system.
In 1299 Metropolitan Maxim moved his residence (the bank's head office) from Kyiv to Vladimir. The chronicle blames Tatar violence in Kyiv. But look at where he went: to the source of the main stream of tribute flowing to the Volga. In 1325 Metropolitan Peter settled in Moscow, with Ivan Kalita, whose nickname means "Moneybag", and laid the foundation of the Dormition Cathedral there. Kalita received the yarlyk for the grand principality only in 1328. So the church reached Moscow before the state title did. It went where the money was going to flow, before the flow actually turned there.
Moscow then did what Beijing and Delhi had done. It did not follow the caravan. It began dragging the caravan to itself. The Horde's collection machine became Moscow's; tribute now went not to Sarai but to the Kremlin; and in 1589 Moscow acquired its own patriarchate (its own local, independent banking system, plugged into the global one).
The second shift was the Baltic. Peter I did not build a trade route; he reached one. The Baltic trade in grain, timber, hemp and tar had existed for centuries, sustaining the Hanseatic League, Novgorod and later Amsterdam. St. Petersburg was founded in 1703, and the court and government moved there in 1712-1713. Here is the key detail showing that both mechanisms worked at once. The capital moved onto the route, and then the route was forcibly bent toward the new capital. From 1713 Peter began restricting trade through Arkhangelsk, the old White Sea port, and in 1721 he ordered foreign ships by decree to sail to the Baltic rather than the north. In 1714 he banned stone construction across the entire country until St. Petersburg was finished. The whole country paid for the new "window to Europe", while the money went west through a single pipe, to Dutch and English buyers.
And the church? In 1721 Peter abolished the patriarchate for good and set up the Holy Synod. The Synod sat in St. Petersburg. Once again the church went where the money went.
Most interesting of all, Moscow had come back once before. In 1728 the court of Peter II moved to Moscow, and the old boyar clans, the Dolgorukovs above all, hoped to keep it there for good. It didn't work. In 1732 Empress Anna moved the capital back to St. Petersburg. The reason was simple: the court lived on exports, and exports went through the Baltic. A capital cut off from the pipe turned out to be a capital without money.
Why the capital went back to Moscow in 1918
Why, then, did the return succeed in 1918, and stick?
The official reason is well known. On February 18, 1918, Germany resumed its offensive, took Pskov within days and then Narva. On February 21 the Bolshevik government, the Council of People's Commissars, proclaimed "The Socialist Fatherland Is in Danger", and on February 26 Lenin drafted the resolution evacuating the government to Moscow. The government train left on March 10 and arrived in Moscow on the evening of March 11; on March 12 a telegram about the move went out to every soviet. On March 16 the Fourth Extraordinary Congress of Soviets approved the transfer. Formally it was temporary: the 1918 constitution of Soviet Russia did not name a capital at all, and only the 1924 constitution officially wrote in Moscow.
But the Germans explain the timing, not why the capital never came back. For that, look at what happened to the Baltic pipe.
By 1914 St. Petersburg had long ceased to be the empire's main trade gateway. Most grain was exported through Black Sea ports, above all Odesa. And the railways, which became the new trade arteries of the 19th century, converged not on St. Petersburg but on Moscow. Moscow became a rail hub where lines from every corner of the country met like spokes. St. Petersburg remained a different kind of window, a financial one. Through it the Russian Empire borrowed from France and Britain and became one of the largest debtors in the world.
Now look at the 1918 calendar. On February 3 the Bolsheviks repudiated all state loans, cutting the financial pipe to Paris and London. In March the capital moved to Moscow. On April 22 came the decree establishing a state monopoly on foreign trade: no cargo could now cross the border without the center's permission. The Baltic was closed by war, Finland and Estonia had broken away, and Petrograd became a border town. The trade pipe was not simply lost. It was shut from the inside.
What was left? What Moscow had started with: collecting tribute from peasants. In May 1918 a food dictatorship was declared, and from January 1919 came prodrazverstka, the forced requisitioning of grain. The armed food detachments that seized grain from peasants were the Horde's levy in new clothes. And for such a machine Moscow is the ideal point: at the center of the radial railways, closer to the grain provinces, far from a sea that no longer delivered anything.
The church, by the way, got there ahead of the state again. A church council restored the patriarchate in Moscow in November 1917, four months before the government moved. The church knows where the caravan will turn, because it remains a banking structure.
What followed only cemented the model. When Soviet industrialization in the 1930s, built by the hands of American firms, needed hard currency, it was raised the familiar way: grain was taken from the peasants and sold abroad. The capital no longer needed to sit on the pipe. The foreign trade monopoly put the tap in Moscow, and whoever holds the tap does not need to live by the port.
So my theory holds, with one fundamental refinement. What comes first is not the caravan as such but the stream of money the elite lives on. If the elite lives on trade, the capital moves onto the trade route: St. Petersburg, Amsterdam, London. If the elite lives on extraction, the capital stays put and drags routes to itself: the Grand Canal to Beijing, the road from Veracruz to Mexico City, Peter's decree against Arkhangelsk, the radial railways to Moscow. And in both cases the church-bank followed the money, sometimes even ran ahead of it.
A test: other countries' moves
The hypothesis can be checked against other countries, and against the very cities on the chart.
In 1911 the British moved the capital of India from Calcutta to Delhi: from a trading port to the old Mughal capital, a classic tax-collection center. Seven years later the Bolsheviks did the same, from Petrograd to Moscow. Others made similar choices. Turkey moved its capital from Istanbul to Ankara in 1923. Pakistan moved from Karachi to Islamabad in the 1960s. Nigeria moved from Lagos to Abuja in 1991.
Now look at the chart. Kolkata still has 20.4 million people inside its circle. Karachi has 21.3 million. Istanbul has 13.4 million, while Ankara has fewer than 6 million. Lagos has 12.3 million. The capital left, the people stayed, because a port lives on the route and does not depend on where the government sits.
Petrograd, by contrast, shrank from roughly 2.4 million people to roughly 720,000 between 1917 and 1920. Today St. Petersburg has about 5.6 million residents and doesn't make the top 30 at all. Where the economy lives on trade, a port keeps its people even without capital status. Where it lives on tribute, money, and therefore work, exists only where the collector sits.
Attali's hearts: rich, but not anthills
The French economist Jacques Attali looks at the history of capitalism through cities rather than states. In every era, he argues, the market order has a "heart" city where money, innovation and decisions concentrate. In "A Brief History of the Future" he describes 9 such hearts and ties each to an innovation. Bruges and the sternpost rudder, 1200-1350. Venice and the caravel, 1350-1500. Antwerp and printing, 1500-1560. Genoa and accounting, 1560-1620. Amsterdam and the fluyt, a cheap cargo ship, 1620-1788. London and the steam engine, 1788-1890. Boston and the piston engine, 1890-1929. New York and the electric motor, 1929-1980. And the ninth heart, Los Angeles and the microchip, from 1980 to the present.
Now lay that list over the chart. Of the nine hearts, only two make the top 30: New York (12.3 million) and London (11.1 million). No Boston, no Los Angeles. Bruges today has about 120,000 residents; historic Venice is down to fewer than 50,000. Genoa, Antwerp and Amsterdam are nowhere near the anthills.
What does that mean? That the heart of capital does not need millions of people within 30 km. It needs a route. All of Attali's hearts are ports. And when the route leaves, the heart does not die; it shrivels to the size of a museum. Here is the first answer to the puzzle of empty Europe: the continent's richest cities were never anthills, because they lived on the flow, not on the territory.
And here you can see how far shifts in trade arteries explain the succession of hearts. Bruges lost its access to the sea when the Zwin inlet silted up. Venice lost when Vasco da Gama opened the route to India around Africa in 1498 and Portuguese spices went to Antwerp instead of through the Levant. Antwerp fell when the Spaniards took it in 1585 and the Dutch closed the mouth of the Scheldt. Genoa rose on American silver, which traveled from Seville via Barcelona to Genoa and on along the "Spanish Road" to the armies in Flanders. Amsterdam grew on the Baltic grain trade, which the Dutch themselves called the "mother trade", and on its East India Company. London seized the Atlantic and India with its navy. And Los Angeles in 1980 marks the turn of world trade toward the Pacific, toward Japan and China.
Nearly every change of heart in Attali's scheme coincides with a change in the direction of the trade artery. The heart of capital really does follow the caravan. And that is exactly why Moscow is not on this list and never could be: it was never a heart, it did not generate capital, it was a parasite, a point for collecting tribute.
Attali and Arrighi: one story at two scales
Now for the most interesting part: how Attali's hearts map onto the systemic cycles of accumulation described by Giovanni Arrighi. Arrighi has four cycles: Genoese, Dutch, British and American. Attali, in effect, breaks them down into cities.
Arrighi's Genoese cycle is not just Genoa. It is an alliance between Genoese bankers and the Spanish crown. The Genoese had money and networks; Spain had the army, the territory and the American silver. And for this same cycle Attali gives two hearts: Antwerp as the market where Portuguese spices and Habsburg finance met, and Genoa as the bank. Attali's date for Genoa, 1560, almost coincides with what Fernand Braudel called "the age of the Genoese", which began with the Spanish bankruptcy of 1557.
And this is where the first prototype of the economics of losses appears. The Spanish crown fought in Flanders on Genoese money. The kingdom declared bankruptcy in 1557, 1560, 1575, 1596, 1607 and 1627. Yet at every restructuring the Genoese received juros, bonds secured on Castile's tax revenues. The losses of a war whose costs could never be verified fell on Castilian taxpayers, while the profits stayed private, in Genoese houses. The scheme I described in my book "The Economics of Losses", published in Ukrainian as "Економіка збитків", was already running in the 16th century, and war was already the most convenient line item.
Arrighi's Dutch cycle corresponds to Amsterdam, the British one to London. Attali splits the American cycle into three hearts: Boston, New York and Los Angeles. The difference in approach is clear. Arrighi describes a regime of accumulation; Attali describes a place. When the route shifts within one regime, Attali moves the heart, while Arrighi keeps the cycle.
There is also a telling coincidence of dates. Los Angeles becomes Attali's heart in 1980. In Arrighi's scheme, that is precisely when the American cycle, having passed through the signal crisis of the 1970s, moves into financial expansion, the autumn of the hegemon. Attali's last move of the heart falls at the start of Arrighi's American autumn.
Why there is no Spain
This also answers why there is not a single Spanish city on the chart. Spain, after all, controlled a whole continent's silver for two centuries. But the silver did not stay in Spain. It sailed into Seville and almost immediately moved on: to Genoese creditors, to Antwerp, to the armies in Flanders. As early as the 17th century people said Spain was like a mouth: it takes in food, chews it and passes it straight on to the other organs.
Spain was neither a heart nor a collector. It was a pipe. Madrid became its capital in 1561 as a court in the middle of Castile, with no sea and no trade route, a pure point of power. But the base it collected from lay across the ocean. When the American colonies broke away in the 1810s and 1820s, Madrid was left as the capital of a country with far fewer subjects than before. A collector with no catchment has nothing to grow an anthill from.
Paris and Germany: a collector without a catchment, a catchment without a collector
Paris is a classic collector. France was for centuries one of the most centralized states in Europe. Taxes were collected by the General Farm, a syndicate of private tax farmers, and the farmers lived in Paris. In 1784-1791 they even ringed the city with the Wall of the Farmers-General, about 24 km long. It was built not for defense but to levy a toll on every cart entering the city. Picture it: a capital surrounded not by fortifications but by a customs wall.
Why, then, is Paris not in the top 30? The Île-de-France region has more than 12 million people, but it stretches far beyond a 30 km circle. Besides, France has only 68 million people and was the first country in Europe to see its birth rate fall. The collector exists, but its catchment is too small for a 20-million circle.
Germany is the opposite case. For centuries it had no single collector. The Holy Roman Empire consisted of hundreds of principalities, free cities and bishoprics, each with its own treasury. Trade ran along the Rhine and through the Hanseatic League, and those routes grew not one city but a chain: Cologne, Frankfurt, Hamburg, Lübeck. That is why even today the country's largest region, Rhine-Ruhr, is spread across a dozen cities. Where the route is rich and the collector is absent, you get a network of prosperous cities, not an anthill.
Vienna: a chance the empire gave and 1918 took away
Vienna, on the other hand, had its chance. And it lost it so visibly that one could not invent a better test of the theory.
In 1910 Vienna had 2.08 million people. It was one of the five largest cities in the world. Why? Because Vienna was the capital of the Habsburgs, who collected taxes from a territory that today belongs to 13 states: from Galicia (now western Ukraine and southern Poland) to Trieste, from Bohemia to Transylvania. Taxes, officials, officers, banks, migrants all flowed to one point on the Danube.
In 1918 the empire collapsed, and the catchment vanished within months. Czechs, Hungarians, Poles and Galicians left for the new states. The Czech community that moved to Czechoslovakia alone numbered more than 100,000 people. By 1934 Vienna's population had fallen to 1.94 million, by 1951 to 1.62 million, and in 1981 it bottomed out at 1.53 million. The city lost nearly a quarter of its population not because of war in its streets but because the territory it collected from had disappeared.
Vienna got back to its 2 million only at the start of the 21st century, once Austria had joined the European Union and the city regained access to a large space. But this time as a hub, not as a collector.
Why Eastern Europe is empty
Now Eastern Europe between the Baltic and the Black Sea. Why is there not a single anthill here?
Because for centuries this land was a catchment from which others collected. In the 16th and 17th centuries Polish and Ukrainian grain went down the Vistula to Gdańsk, and from there to Amsterdam. This was the very trade the Dutch called their "mother trade". The Polish nobility lived on grain exports, and for that it needed not cities but estates and enserfed peasants. That is how the "second serfdom" arose. The heart of capital sat in Amsterdam, and Eastern Europe was its field.
Then came the partitions of the Polish-Lithuanian Commonwealth, and the taxes from these lands went to St. Petersburg, Vienna and Berlin. Then the 20th century: two world wars, the Holocaust, deportations, famine. The region lost millions of its urban population. And after 1945 the tribute flowed east again, to Moscow.
Eastern Europe never had a collector of its own that kept what it gathered at home. The collectors always sat beyond its borders. And an anthill grows not where things are produced, but where they are collected.
Did Kyiv have a chance?
Which brings me to a question I cannot avoid as a Kyiv resident: did my hometown ever have a chance to become a megacity?
Yes. And precisely when Kyiv stood on a trade route. In the 11th century Adam of Bremen called Kyiv a rival of Constantinople, and Thietmar of Merseburg wrote of its 400 churches and 8 markets. Kyiv was the heart of its route "from the Varangians to the Greeks": not an anthill, but a rich trading city, as Bruges and Venice would later be. And it had its own church bank: the Kyiv Monastery of the Caves took deposits from princes and lent money.
But when the route dried up, Kyiv shared the fate of Bruges. 1169, 1203, 1240: three sackings in barely seventy years. The metropolitan followed the money to the northeast. After that Kyiv spent centuries as a province of other people's collectors: Lithuanian, Polish, Muscovite. Even in Soviet times Moscow kept the capital of Soviet Ukraine in Kharkiv, closer to itself, until 1934. Taxes, grain and people from Ukrainian lands went not to Kyiv but through it.
So Kyiv never had a chance to become an anthill through tribute, because whatever was collected from its land always drained away to someone else's capital. Its chance was of another kind: the chance of a trading heart. And that kind of chance is measured not in millions inside a circle, but in whether a living route runs through the city.
Africa: a continent carved into pumps
All of Africa, more than 1.4 billion people, gives the top 30 just 4 cities: Cairo (25.3 million), Kinshasa (16.3 million), Lagos (12.3 million) and Luanda (11.5 million). Why so few?
Look at who these four are. Cairo is the oldest agrarian collector on the planet. Egypt has more than 100 million people, and nearly all of them live in a narrow strip along the Nile, because the rest of the country is desert. Cairo stands in the middle of its own grain, like Lahore. The other three are pumps. Kinshasa was founded by Henry Morton Stanley in 1881 as Léopoldville, a station of the Belgian King Leopold II on the Congo River, through which rubber was extracted. Lagos was a British port that shipped slaves, then palm oil. Luanda was a Portuguese port for exporting slaves, and now exports oil.
Why are there no others? Because at the Berlin Conference of 1884-1885 the continent was carved up among European powers, and those possessions later fragmented into more than 50 countries. Each colony had its own small pump, a port-capital through which its resources flowed to the mother country. No single point collected from the whole continent. Africa had not one collector but dozens of small pumps. A divided catchment produces many mid-sized cities and no giants, except where the pump stood on the biggest river or the richest oil.
The Middle East: the route is there, but it no longer stops
Now the Middle East. Are there no trade routes there? There are, and they are among the most important in the world. But look at what happened to them.
The region once held the largest cities on the planet. Baghdad, founded in 762, was by some estimates the largest city in the world in the 9th century. It sat at the crossroads of caravan routes linking India, Persia and the Mediterranean, while the caliphate collected tribute from a vast territory. Both mechanisms worked here at once. In 1258 Hulagu's Mongols destroyed Baghdad, and the city did not regain that level for centuries. Cairo in the 14th century was the largest city of the Mediterranean, because spices from India traveled to Venice through the Red Sea and Cairo.
Then the route simply went around the region. In 1498 Vasco da Gama opened the sea route around Africa, and spices bypassed the Middle East. In 1517 the Ottomans conquered Egypt, and tribute now went to Istanbul. That is why Istanbul is still in the top 30: it sat both on the route through the Bosporus and on the tribute of an entire empire.
After 1918 the Middle East was carved into mandates under the Sykes-Picot agreement of 1916, and separate states grew out of them. Same as in Africa: instead of one collector, several dozen small ones.
Modern routes, meanwhile, work differently. Since 1869 the Suez Canal has carried ships in transit: a container ship sails from Shanghai to Rotterdam and does not stop in Cairo. Oil moves by pipeline and tanker, and oil extraction needs capital, not millions of workers. The rent goes to a narrow circle of citizens, while the work is done by migrants who come and go. So the region's oil and transit cities end up rich but not anthills: Dubai has about 4 million people, not 20.
The answer to "are there no routes there?" is therefore this: the route is there, but the modern route no longer stops. And an anthill grows only where the flow is stopped and unloaded.
Why London and New York are anthills
So back to London and New York. Why did they not shrivel, as Bruges and Venice did, after the heart moved on?
Because both became the financial centers of their hegemonies in the autumn phase. In Arrighi's account, when a hegemon's material expansion runs out, capital leaves production and trade for finance. For Britain this happened after the 1870s, for the United States after the 1970s. At that moment the heart of capital changes its nature. It no longer stands on a trade route. It collects rent from the whole world, through credit, currency, the stock exchange, insurance. The City and Wall Street became the points where the fee for using other people's money flows in.
And here Arrighi meets the economics of losses. In 2008 the losses of Wall Street banks were nationalized through the $700 billion TARP program, and the British government rescued its own banks the same way. The profits of earlier years stayed private; the losses were shifted onto taxpayers. That is the very mechanism that prolongs the hegemon's autumn: the state takes on the risk of the financial center, and the center lives on.
But 2008 also showed that this mechanism is not a law of nature. It is a choice. Two small island nations went through the same crisis by opposite paths.
On September 30, 2008, Ireland guaranteed the liabilities of six of its banks. The guarantee covered around €400 billion, more than twice the country's annual economic output. Two years later the guarantee had to be honored. In November 2010 Ireland took an €85 billion rescue package, roughly $111 billion, from the EU and the IMF. Saving the banks cost Irish taxpayers €64 billion, and the country sank for years into austerity and public debt.
Iceland did the opposite. Its bankrupt Landsbanki had used an online brand, Icesave, to collect deposits from hundreds of thousands of Britons and Dutch savers. The British and Dutch governments paid their depositors €3.9 billion, about $5.5 billion, and demanded that Iceland repay them at its taxpayers' expense. Icelanders said "no" twice in referendums. On March 6, 2010, 93% voted against the deal; on April 9, 2011, 59% did. The case went to the EFTA Court, which ruled on January 28, 2013 that Iceland was not obliged to pay.
A private bank went bust, and the taxpayers refused to take its losses onto themselves. The economics of losses works only as long as society agrees to pay.
London and New York agreed. And that is why, in the autumn phase, they became what Moscow had been from birth: a point where an elite concentrates resources gathered from a territory far larger than the city itself. The only difference is that London and New York collect their tribute through finance, and Moscow through force.
Collector and periphery today
Moscow shows the model in its purest form. Mayor Sergei Sobyanin himself states on the Moscow city government's website that the economy of Moscow and its surrounding region makes up almost a quarter of Russia's entire economy. Not because Moscow produces that much, but because the headquarters of the commodity companies through which the oil and gas rent flows are registered in Moscow, and that is where the taxes land. The oil is pumped in Siberia; the reporting and the tax payments happen in Moscow.
Now the other half of the balance sheet. According to counts by Mediazona, Meduza and the BBC Russian Service, the republic of Tuva loses 120 dead in the war for every 10,000 men, Buryatia 91, the Altai Republic 89. In Moscow the figure is 3, as The Insider reported. A 40-fold gap.
This is the economics of losses in spatial form. Risk and death are pushed out to the periphery; money, contracts and budgets stay at the center. War is the most convenient line item here, because its cost cannot be verified, and those who pay for it are invisible from the center. In my book "The Russian Myth", published in Ukrainian as "Русскій Міфъ", I argued that Russia started this war to write off $640 billion in reserves that could not be returned into the economy. The mechanism is the same as in 16th-century Castile, except that instead of Castilian taxpayers it is the Buryats and Tuvans who pay.
The first Rome and the second
One last question remains: why Rome did not become a megacity. It was Rome, after all, that first gathered a million people in Europe.
Ancient Rome was a pure collector. It barely produced anything, and barely traded what little it did. It ate grain shipped in from Egypt, Sicily and North Africa as tax. The state grain dole, the annona, fed hundreds of thousands of Romans. A city of a million stood on the tribute of the entire Mediterranean.
In 330 Emperor Constantine founded a new capital on the Bosporus, and within a few years the Egyptian grain was going not to Rome but to Constantinople. The tribute was simply rerouted. Then came the Gothic wars, the Vandals, the loss of North Africa. Rome slid to a few tens of thousands of inhabitants.
The church-bank lifted it once more. The papal curia collected tithes, annates, Peter's Pence and indulgence money from all of Catholic Europe, and that money settled in Rome. But look at what happened when the popes moved to Avignon in 1309-1377. The bank left in search of safety and money, and Rome, abandoned, shrank to a few tens of thousands of people. Again the church left first, and the city merely followed. Then, in 1517, Luther came out against the sale of indulgences, and the Reformation cut northern Europe off from the Roman till. In 1534 England stopped paying Peter's Pence. The catchment shrank sharply, and Rome never became an anthill again. Today its entire metro area has about 4.3 million people.
Now look at Constantinople. It received the same Egyptian grain, but it sat on more than tribute. It sat on the Bosporus, on the route between the Black Sea and the Mediterranean. Byzantium fell, the tribute changed hands, but the route remained. Today Istanbul has 13.4 million people inside its 30 km circle.
The first Rome lived on tribute and died with it. The second Rome stood on a route and outlived its own empire by more than 500 years. And Moscow, which proclaimed itself the Third Rome, chose the model of the first Rome, not of Constantinople. Its 15.3 million rest on tribute from a vast territory, and the chart shows exactly how big that tribute is.