A long tradition claims that one factor that distinguished western Europe from China, the Middle East, and Russia was the presence of independent city-states. Max Weber, Henri Pirenne, and John Hicks argued that city-states played a crucial role in beginning the long road to modern economic growth (see this earlier MR post on producer and consumer cities).
De Long and Shleifer (1993) argued that city-states ruled by merchants favored policies that protected property rights and markets and that they imposed lower taxes than princely states did. But historians have found that cities like Florence actually imposed much higher taxes than feudal states did (see Epstein 1991, Epstein 1993 [JSTOR links], or Epstein 2000). Until now, however, no-one (to the best of my knowledge) has provided systematic evidence on the economic performance of independent city-states across Europe. A new paper by David Stasavage attempts to do just this. Using city population as a proxy for economic development, he finds that autonomous city-states overall didn’t grow faster than other cities. Interestingly, new city-states which had been independent for less than 200 years did grow faster. After more than 200 years of independence, growth in these independent cities slowed and they grew more slowly than did ordinary cities.
Stasavage interprets this finding in terms of a model of oligarchies proposed by Acemoglu (2008). Oligarchies initially have an incentive to impose institutions that favor markets and economic growth. However, oligarchies also impose barriers to entry, and over time these barriers to entry lead to growth slowing down. Other (complementary) mechanisms may also have been at work. In particular, Stasavage does not consider the role of external warfare. Once they became rich and prosperous, city-states were often attacked by neighboring territorial states. Many city-states then had to impose high taxes and other extractive policies in order to survive. In any case, these findings are significant for an argument that I will evaluate in subsequent posts that claims that the rise of state capacity played an important role in getting growth going in early modern Europe.
How would it take 200 years for oligarchies to establish themselves in medieval cities? This would make sense if the gap were one or two generations, but not seven. Warfare does make more sense as an explanation.
You forgot the initial growth period.
So maybe it’s catch-up growth and then decay: 100 years in a medieval context to catch up under the new free market overlords, and then 100 years for rent-seeking to set in and definitively strangle further growth.
It seems to me that while population growth is a good proxy for a Malthusian Agricultural model of economic growth dominant through most of the pre-industrialized world. However, urban economies which city states largely relied on were more capital intensive and should be modeled using a Solow Model where population growth would be a worse measure of growth. My general impression is that there were many population control methods used in the pre-modern world and I suspect manufacturing and trade intensive city states would have used them as there were lower returns to childbearing. If they showed strong correlations in the city states between population growth and signs of a well fed population like bone density, I think this would be a far more convincing argument.
MikeJ
Migration?
ComplEmentary, dude.
Regarding the military story: surely in the grand scheme of things, military (and other kinds of) competition between states is a big reason for European technological progress? That’s what Gibbon thought, anyway.
Be careful when conceptualizing competition in terms of destructive activity – a lot of economic intuition goes haywire when you’re allowed to bop your neighbor across the head and nick his stuff. You can easily get a bad equilibrium where physical technology is insufficient to overcome managerial innovations in organizing hordes of nomadic horsemen, as in did in fact happen in Central Asia.
OTOH historically, transitions from a bad equilibrium to a better one have mostly been quite destructive and violent.
20th century “growth miracles” Israel, South Korea, and Taiwan had to deal with quite a bit of military competition as well, though there were several other things going on.
US aid, US aid, US aid.
Interestingly it took 100-200 years for the fledgling US government to become significantly more extractive.
Conclusion: After 200 years the only productive step is to break the entire country apart and start again?
this doesn’t quite address eg Paul Kennedy’s view, which i don’t think was ever that the city-states did better than non-independent cities, but that the presence of many competing city-states/oligarchies in constant competition with one another provided the impetus for continued innovation and development compared to places like ming china or mughal india where single rule over huge swaths of territory went unchallenged for long periods of time and the ruling classes essentially sat on their laurels rather than promote innovation or even outlawed innovative developments as they had no need for them (ming).
aside from Venice, the other city states were run over by either France, Spain or the Austrian branch of the Hapsburgs.
Mughal india may have had one political unit, but Southern India and the surrounding states had a huge number of separate states to rival the conditions in Europe.
Multiple competing states may be necessary, but they are not a sufficient condition.
I guess that city states are beneficial mostly when they are immersed into a city state culture as whole. Thus each city competes with the others and as result we would have a process of institutional competition and the gradual development of better institutions.
Apparently, it was in antiquity the height of city states, when there were 1,000 – 1,100 city states in the Mediterranean focused on the Aegean, that there was radical institutional innovation. And, given the scarcity of evidence from the distant past, it appears that there was some very significant economic growth during the period as well (http://www.princeton.edu/~pswpc/pdfs/ober/051001.pdf).
Also, our modern world is almost a “city state world” since there are 200 countries in the world, which is a greater number than the number of cities of 10,000 inhabitants in 16th century Europe, though each country is 30 million on average, given the high degree of integration in our world today, they more alike city states than medieval/early modern empires, which existed in political systems with one or few states (such as the world empires of Ming China and Mughal India).
But the period of radical innovation in the Mediterranean city states was pretty brief if you associate their beginings with the revolutions that marked the end of the tyrannies around 550 BC and their destruction by Alexander and his Generals in the late 300s. And if you look at Italy and Rome, note that Roman very quickly became a territorial state, by the end of the Second Punic War around 200 BC, it was by no means a city state. I would even argue that Rome ceased to be a city state at the time of Pyrrhus in the 270s. If you factor in 200 years you take Rome back to about 450 you are only about fifty years off from the expulsion of the Tarquins in the traditional date of 509 BC.
The 200 year thing holds up really well.
Wow, this really chimes with my understanding of Italian City States. Once the wars between the Ghibbilenes and Guelphs were basically over they had a brief florecence that ended before the French invasion. In Germany the city states died about the same time as innovators, and the Guild system quickly became one of the most reactionary forces in European civilization. But what makes this even more interesting is that if you scale it up to leagues of city states you see the same process occur, only more rapidly, in the United Provinces, which were in the forefront of commercial development for about 150 years and then quickly become stultifying and decadent before he beginning of the 18th century. In all three cases, Italy, Germany, and the Dutch State, they went from being at the vanguard of development to being commercial has beens, and then showed a remarkable sticking power at economic and social stasis. In Italy this has lasted until today, note that Milan was only very briefly a city state, in Germany this lasted until the end of the nineteenth century with the centralization of the German Empire after Bismarck was dismissed (if you doubt this look at the evolution of the Free cities, such as Bremen, Hamburg, and Lubeck in the late 19th century), and the Netherlands which was conquered by the Revolutionary French which in the period between the 1790s and the 1830s, with the independence of Belgium, was completely and utterly transformed. This would also nicely conform to what we understand history of the Mahajanapadas of Guatama Buddha’s time, and I could probably relate it to the evolution of the Greek city states from the time of the overthrow of the tyrants to the 4th century and their extinction as independent entities.
As to why it would take 200 years to consolidate power in a city state, think about plague, war, and the social revolution involved in founding one. I think I really need to read Acemoglu.
City states being major differentiators between China and Europe does not pass the smell test. Even if they were bastions of free trade I don’t think more than 5% of Europe’s population at any given time lived in city states. So even if they grew red hot it would still be a negligible contribution to Europe’s overall higher pre-industrial growth.
…..unless most of the economic growth came from this 5% of the population. Not hard to imagine; productivity is always very asymmetrically distributed.
Using city population as a proxy for economic development,
Since when is population a credible indicator of development? That assumption sounds bizarre.
Planet Earth?
For preindustrial cities it isn’t really that bizarre. Mortality rates in urban areas were high enough that population growth was rarely endogenous – cities grew only to the extent that people in rural areas would look at them and think, “Hey, my odds of making a fortune and retiring to a country estate before I die of the plague, look better in this city here than that one over there!”
Most of these people will in fact die of the plague before they ever make their fortune, but so long as the economic development of a city outpaces its population growth it will still attract its share of optimists. Population outgrows economy, and the immigrants stop coming.
“A long tradition claims that one factor that distinguished western Europe from China, the Middle East, and Russia was the presence of independent city-states”
An odd thing about this claim is that it’s (obviously) false about Russia (at least- I don’t know about the others.) Kropotkin uses the Russian medieval free cities as one of his prime examples in _Mutual Aid_, after all. (Pskov is a major example, but not the only one.) They didn’t survive long into the modern era, but their demise was for reasons quite similar to that of city-states at other times and places. So, I’d think this account doesn’t even get off the ground, since it starts from a false presumption.
Actually, I was thinking about caveating the point about Russia by mentioning the Novgorod Republic. I didn’t know much about Pskov. But the point is that Muscovy made short work of these cities whereas in western Europe city states survived for centuries.
Exactly. Unlike Western Europe the medieval city states did not survive as even semi-independent entities in Russia. Novgorod, for example, was decimated by Muscovy, even its dialect of North Slavic went extinct.
I’m really liking these economic history posts by MKoyama. Anyone have a good list of other blogs that discuss similar things? I get enough about the here-and-now economic policy, I want to know what’s already happened in the past…
Well, except for the fact that medieval, pre-Mongol Russia consisted of city states, albeit they controlled more land around them than, say, the city states of ancient Greece. But Kiev, Ryazan, Novgorod, Vladimir-Suzdal, etc were all independent states, generally with a form of mixed republic-limited monarchy form of governance– early Russian princes were no more absolute autocrats than early European monarchs more, and in fact they could be legally deposed as even Alexandr Nevsky found out when he got a bit too high and mighty for the burgers of Novgorod.
Before they were subjugated under the Tartar Yoke, the Russian city states made huge progress, look at the way the Swedes and initial waves of Turkic peoples were driven back and the creation of highly functioning urban polities in an area that had never supported actual cities prviously. They were hugely productive until a premature stasis was forced on them by the Mongol conquests.
Friends, I may be late to the party. Because of an obsession with city-states, I looked at the paper and noticed that, page 19 the equation growth = alpha +beta (autonomy)+…+eit does not hold under fat tails. I don’t know if his argument stands, but as it is, adding table 3 (and others) “standard deviation” and it does not look like the article means anything. The results can be killed because he uses the wrong norm L2 with anyway weak results.
In other words, whatever math he uses does not lead to his conclusion.
I am offline but if someone would like to comment please send me an email and I will reply.
Happy father’s day,
Nassim
Is it just that it took 200 years to converge to a new, higher steady state (relative to the economic frontier)? That would leave institutional decay out of it.
Comments on this entry are closed.