The Rise and Fall of the Ancient Economy

I have already posted the information of this thread in my blog, but due to the limitations of the number of pictures that I can put on blog posts I decided also to post the data in thread format.

Using the products of modern archaeology it is becoming possible for us to reconstruct the trajectory of the total volume of economic activity of the classical world during antiquity. The evidence apparently indicates that aggregate economic activity in the ancient world peaked in the 1st century AD, after rising for nearly a thousand years and then started to decline, nearly continuously, for nearly a thousand years.

Metallurgy

Recent research using the levels of metal pollution in the ice cores from greenland is enabling researchers to determine the trajectories of pollution and hence production of several metals, allowing us to reconstruct the performance of several industries of the Graeco-Roman world.

Lead

Measured from levels of lead pollution, world production of lead increased continuously from the discovery of cupellation up to the height of the Roman Empire:

(1)

Each dot represents the point estimates of lead production in the world while the outputs in other periods were derived from interpolation. The graph was made in logarithmic scale, which fails to show the massive difference in levels of lead production of the Roman period to other periods, so I converted this data to linear scale:



Lead production levels during Roman times reached levels attained only again in the 18th century. After rising continuously from 3000 BC, lead production started to decline from the 1st century AD, failing from 80,000 tons to 4,000 tons in the 8th century AD. Production took a thousand years to recover to former Roman levels, reaching 100,000 tons in the late 18th century.

Lead was a by-product of silver smelting. The growing classical economy demanded increasing quantities of coinage, which increased the demand for silver and thus increasing the production of silver and it's by products. With the decline of the ancient economy, starting in the 2nd century, silver production also declined and thus the levels of lead production. So lead pollution can be viewed as a thermometer of overall level of economic activity in the ancient world.

Average Greco-Roman silver production in tons per year

350-250 BC -------- 25
250-150 BC -------- 60
150-50 BC -------- 100
50 BC-100 AD ----- 200
100-200 AD ------- 100
200-300 AD ------- 30
300-400 AD ------- 25
(5)

Roman levels of silver production of the 1st century were only surpassed in the late 16th century, with the opening of the mines in the new world. Yet, in the mid 18th century, world levels of silver production were in the order of 600 tons, three times Roman levels, but dispersed across a much larger population, so Roman per capita levels of silver consumption were higher.

Lead was also used to manufacture pipes in the ancient world, in Pompeii many lead pipes running though the streets have been found.

Roman lead ingots:



Copper

Like lead, the global trajectory of copper production followed a similar path: increasing continuously from the discovery of copper smelting to the height of the Roman empire, then it started to fall and reached it's lowest levels in the 8th century, starting to increase again and finally surpassing the Roman levels in the late 18th century. Though with one major difference, which was the economic boom in Song China during the 10th - 12th centuries, that resulted into a medieval peak in copper production similar to the Roman and 18th century levels.

World copper production (logarithmic scale):

(2)

And here is a table detailing the evolution of copper emissions:

(2)

Note: from 1000 BC to 1 AD world copper emissions increased from 20 tons a year, to 300 tons a year in 500 BC to 2,300 tons a year in 1 AD. Increasing by a over 100 times, showing clearly the rise of the ancient economy during the period. From 1 AD to 750 AD world copper emissions declined from 2,300 tons to 300 tons, back to the levels of 500 BC, a decline of 87%, smaller than the decline in lead emissions/production of 95%. The difference can be explained by the fact that China was a significant producer of copper, while they didn't produce lead in significant quantities. Thus, mediterranean levels of copper production declined by an even greater margin than 87% from the 1st to the 8th centuries. And also, in 500 BC, the main center of copper production was centered in the mediterranean (Hellas and the Persian Empire), while in 750 AD, China probably produced more copper than the western world. So actual levels of copper production in the western world were lower in 750 AD than in 500 BC.

Iron

Iron was the most widely used metal in the Classical Mediterranean, however, since iron production doesn't release special elements in the atmosphere we cannot determine clearly the trajectory of iron production during ancient times. However, based on local archaeological evidence, we can reconstruct the trajectory of iron production in certain areas of the ancient world.

In Britain, where extensive surveys of iron making sites were made, we can roughly reconstruct the trajectory of iron production. Britain was inhabited by celtic tribes when it was conquered by Rome in the mid 1st century AD, a time when the economy of the classical mediterranean was at it's peak prosperity. As result, following the conquest, the province was developed very quickly. Road and cities quickly emerged in the island. While iron making sites increased exponentially, in the case of this region, from 1 iron making site before the conquest to dozens after:


(6)

From the discovered sites, a rough estimate of the trajectory of British iron production can be constructed:



(6)

The British iron industry was very small if compared to the scale of silver mining, at about 1,500 workers at it's peak production. Athens had 20,000 slaves working in the Laurion silver mines, while Roman Iberia probably had in the order of well over 100,000 workers working in the silver and gold mines: the whole Roman silver industry is estimated to have employed about 200,000 workers.

Notice also the dramatic decline of the estimated levels of iron production from 200 AD to 400 AD, from 1,740 tons in the early 3rd century to 270 tons in the late 4th century. Reflecting the dramatic decline of the Mediterranean economy over the period.

More modern estimates (7) puts British iron production at about 2,250 tons, considering that Britain had about 2-2.5% of the population of the Empire (4), it would imply in a total imperial production of ca. 100,000 tons. For comparison, Europe produced 180,000 tons of iron in 1700 AD, at a similar level in per capita terms (the population of the Roman Empire was ca. 65-70 million while Europe in 1700 had 120 million inhabitants). Though, relative to the levels of copper and lead production, this estimate appears conservative: world copper production only reached Roman levels in 1800 AD, when iron production was 1 million tons. I suspect that Roman iron production levels were in the order of hundreds of thousands of tons, tanking into account the fact that iron was cheap enough in the Roman Empire for the banding of wagon wheels, which became standard again in Europe only in the 19th century. (9)

Shipwrecks

Another indicator of the relative levels of economic activity in the mediterranean region are dated shipwrecks. The mediterranean sea had a huge importance for the Classical economy by providing a true highway for the transport of bulk commodities:


(8)

Transport by land was ineficient and costly. Economic growth depends on the division of labor and division of labor depends on trade. Trade in turn depends on the costs of transportation. And maritime transport in Roman times was 50-60 times cheaper than transport by land. As result there were 180,000 kilometers of trade routes across the classical Mediterranean:



By using a dataset of 1,200 dated shipwrecks from 2500 BC to 1500 AD we can get a rough idea of the relative levels of volume of trade in the Mediterranean sea over the centuries:



Notice how it mimics the trajectory of copper and lead production: a 100 fold increase during the 1st millennium BC and a 100 fold decrease from the 1st century AD to the 8th century AD. And notice the recovery during the late middle ages, thanks to the Italian Cities. However, the volume of trade probably didn't reach the same magnitudes as in Roman times, simply because it involved a much smaller proportion of the Mediterranean: in Roman times the whole Mediterranean was a network of city states trading with each other. To give an idea, according to Maddison (10), the merchant fleets of the 15th century mediterranean had 220,000 tons of carrying capacity, while Roman Egypt alone already exported 140,000 tons of wheat to a single city of the Roman Empire, 1,400 years earlier. The total volume of trade over long distances was probably in the range of several million tons.

Shipwrecks distributed at a 20-year resolution, now the law of greater numbers still didn't act enough to allow a sharp resolution on the trajectory of maritime trade, we know it was highest in the 100 BC to 100 AD period, but not for each 20-year slice of time, we would need more than 1,200 shipwrecks to allow a greater resolution analysis of the shipwreck data:



Here is the distribution of shipwrecks across the mediterranean:


About 70% of the shipwrecks were from the Western Mediterranean while 30% were from the Eastern Mediterranean, this reflects the fact that the Western Mediterranean consists of the shores of first world countries while the Eastern Mediterranean countries are poorer today and thus cannot afford as much diving (many shipwrecks are discovered by tourists, who stumble upon the shipwreck accidentally)

Greco-Roman shipwrecks were not only much more numerous than in earlier and later periods, but also they were larger: Ships in the bronze age and in the middle ages were usually of dozens of tons. While Hellenistic and Roman period shipwrecks were larger, usually 70 to 120 tons and many of several hundreds of tons (400-500 tons). The largest ships were the grain carriers, though to have been of 1,000 - 1,300 tons, however, and these ships are invisible in the archaeological record because they didn't carry pottery containers which make discovery and the identification of the wreck much easier.

Also shipwrecks and lead pollution are closely correlated because both show the same thing: general levels of market activity, measured in the volume of maritime trade and in the demand for money/coinage and it's impact on the scale of the silver industry and it's lead by-product.



The mediterranean was the lifeblood of classical civilization by allowing long distance trade of bulk commodities to take place in a large scale, perhaps only in the 19th century, with the invention of the railroad, that the European economy attained and surpassed the levels of integration that the mediterranean economy had achieved two millennia earlier.

Animal bones

Here is the distribution of mammal bones dug in the provinces of the Roman Empire:


(4)

Mamal bones are an indicator of economic development because they indicate the size of the cattle/pig/sheep livestock of the Roman Empire, and thus the overall levels of meat consumption. Rich people consume more meat and less grain than poor people, thus this is an indicator of the likely total number of people that consumed meat. Note that in Classical Greece meat consumption was done on a smaller scale, but with the conquest of northern Europe and massive grazing lands, the Romans now could enjoy large quantities of meat.

Note that like shipwrecks, most archaeological work done in the provinces of the Roman Empire refers to the Western European provinces and thus, to Gaul, Iberia and Britain, areas that the Romans conquered from "barbarian" tribes and thus areas that weren't developed before conquest. Explaining why animal bones increased only later, in the 1st century BC onwards.

Population density

Extensive field survey research is enabling us to have a good idea of the overall trajectory of population density. Here is the result of field surveys conducted on the Albegna valley in central Italy, a core region of the mediterranean:


(3)

Population densities reached their peak in the 50 BC to 100 AD period, in other sites the same pattern emerges:


(3)

Population density peaks just where the shipwrecks, lead, copper data peaks as well. Just as we expect from the trends we have been seeing over the last 20 graphs.

Note that estimated population levels from field surveys are ultimately the product of things such as pottery fragments and (in case of the urban population in the Albegna valley) walled area, but these factors are perhaps also a product of the size of physical economic activity (pottery production and ruins) rather than only of population levels. They show that the Italian countryside enjoyed it's peak levels of prosperity during the period from ca 50 BC to the early 2d century. In Greece, the estimated population of sites peaked earlier in the 480 to 200 BC (which also implies that physical economic activity peaked in this period in the Aegean):

Koressos on Kea

900 BC to 700 BC - 100
480 BC to 323 BC - 1,020-1,455 (increase of 10.2x - 14.55x)

Southern Argolid

750 - 650 BC - 1,100
480 - 200 BC - 10,855 (increase of 9.87x)

Source: (12)

Reflecting that the mediterranean economy and population didn't follow the exact same trajectory in all regions. While aggregate levels of economic activity peaked in the 100 BC - 150 AD period, in some regions economic activity peaked earlier (particularly the Aegean and probably the rest of the Eastern Mediterranean enjoyed peak levels of economic activity in the late Hellenistic period), while in other regions economic activity may have peaked even later, such as in Roman Britain, whose levels of economic activity peaked in the early 3rd century, if we can trust the quality of those estimates of iron production and it's correlation with overall activity.

The Classical World at it's Zenith

At it's zenith in the late 1nd millennium BC and the early 1st millennium AD, the Mediterranean economy became so prosperous that demand for luxury goods increased to the extent of creating a general flow of goods from the most distant exotic places in the world, such as India, China and the Arabias, into the Mediterranean basin. Also, the spread of Roman mines into northern Europe, such as the Dacian goldmines, reflected the same increasing demands of the Mediterranean basin. The silk road was opened in the late 1st millennium BC due to the growing consumer market of the Mediterranean world, and as result, the Mediterranean economy had an indirect effect over perhaps 80% of the world's population.

The Mediterranean economy was integrated to the extent that prices of grain reflected it's distance to the city of Rome:


(12)

And a single workshop in Gaul produced pottery that was distributed over thousands of kilometers across the empires:



The high level of productivity in the mediterranean economy was also reflected into the patterns of consumption. Pompeii was a typical Italian town of the 1st century mediterranean world, however it's standards of consumption were much higher than those of other ancient and medieval cities.

A typical 1st century house. Houses in 1st century Roman Italy were huge and very nicely decoared, they had on average 7 to 8 rooms at the first floor and 250-260 square meters (18):


From the Cambridge Economic history of the Greco-Roman world:

(4)

According to the historian, W. Jongman:

Jongman said:

With the growth of it's Empire, with the growth of it's cities and with the growth of a government and a system of transportation based on those cities, Rome had created perhaps the most prosperous and successful pre-industrial economy in history. The age of Antoninus Pius was indeed probably the best age to live in pre-industrial history.

Click to expand...

(14)

Though I cannot fully agree with this assertion. While in aggregate terms it appears that the Early Roman Empire was the most prosperous period in ancient and medieval history, in terms of living standards for the general population Classical Greece appears to have been significantly more prosperous than the Roman Empire in the average, though the most developed regions of the Roman Empire were perhaps on the same level as the Greek cities 400 years earlier. Also, the 17th and 18th century Netherlands were probably wealthier in average than the Roman Empire, though I wouldn't say it was definitely wealthier than the most advanced regions in the empire. In terms of wages we have the following evidence:

Place ------ Daily wages in kg of wheat

Athens 350 BC -------------- 11.5
Preatorian 1st century ------- 10.1 **
Antwerp 1775 AD ------------ 9.6
Pompeii 79 AD ---------------- 8.7 **
Legionary 1st century -------- 7.9 **
England 1775 AD ------------- 7.0
Auxiliary 1st century ---------- 6.6 **
Paris 1775 AD ---------------- 5.9
Roman Egypt 2nd century ----- 5.0
Madrid 1775 AD -------------- 4.8
Florence 1775 AD ------------ 3.3
India 1775 AD ---------------- 2.3 *
Japan 1775 AD --------------- 2.3 *
Dacia 2nd century ------------ 2.2 **

* Converted from rice in basis of calorific equivalence, but note that wheat has more protein than rice per calorie, so this measure actually overstates Japanese and Indian wages.
** Wages in Pompei were 1 denarii a day, while wheat prices in the provinces varied from 3 to 2 sestertii per modii, from 2 sestertii in Egypt and Palestine, 2.5 sestertii in Syria and 3 sestertii in the other provinces, while wheat prices in the wheat sinkhole of Rome were 6 to 8 sestertii per modii. For Pompeii and Legionary, Auxiliary, Dacian wages I assumed 3 sestertii per modii, for Preatorian wages, I assumed 7 sestertii per modii, for Egypt I assumed 2 sestertii per modii. For converting soldiers annual stipends into daily wages I assumed 250 working days per year, a standard number of working days for Early Modern Europe, while (17) claims that in ancient Greece it appears that 200 to 250 days was the standard number of working days per year.

Sources:
Roman wages: (3)
For the 18th century wages: (15), (16) for Japan

Notice, Roman wheat wages were similar to the late 18th century world, and varied by a large measure, from 10.1 kg per day to the soldiers in the Preatorian guard, a force of about 10 - 15 thousand men in the city of Rome, which had annual salaries 3 times the salaries of the standard Legionaries, a function of the much higher cost of living in Rome compared to the rest of the empire: Grain prices in Rome were 6 to 8 sestertii per modii, while they were usually 2 to 3 sestertii outside Rome.

The Decline of the Classical Economy


By the early 1st millennium AD the civilization of the ancients reached it's peak levels of prosperity (in aggregate). Over the following centuries the long run trend was only of decline, while there were some shorter periods of prosperity in some regions, the overall trend was rather clear, as we have seem in the case of lead, copper, animal bones, British iron production and shipwrecks: a large decline followed. I also have the following pieces of data:


(14)

This graph refers to the heights of individuals, hence to the biological standards of living, a reflection of the levels of nutrition achieved by Roman Italian population, they show a peak in the late 1st century, just when all other data peak as well, and later a not very clear cut, but still visible, trend of decline in heights. Also reflected in the reduced consumption of meat and milk, the products of the cattle livestock, which declined in size as we have already seem.

The size not of people but also of the animals such as cattle also declined with the decline of the mediterranean economy, reflecting the decline of the cattle raising industry: from a professional industry that produce cattle for large scale consumption of food to a local activity performed in a small scale in the early medieval villages.


(13)

The decline is also apparent in the raw number of archaeological objects excavated from Western Germany, an area that the Romans conquered in the 1st century BC and which reached peak prosperity in the late 1st century AD, like the rest of the empire:


(14)

According to the MIT economist P. Temin:

Temin said:

Even though the Roman economy differed from early modern European economies in many respects, it appears more similar to our economy than the Medieval economy. Large-scale production and movements of resources were dominated by markets, although redistribution and reciprocity were present as well. This mixture of modes is present even today; the important characteristic of the Roman economy was its market orientation. This mode of organization promoted trade and the exploitation of comparative advantage, helped by political stability, personal security, and widespread education. All this undoubtedly contributed to the prosperity of the early Roman Empire, at least for the upper classes, which was not to be equaled in the West for almost two millennia thereafter.

Inflation accelerated in the third century CE, visible to us through debased coinage and occasional price quotations (Harl, 1996; Rathbone, 1997). Prices may have doubled in the second century, possibly in a discrete jump after the Antonine Plague in the late second century; the rate of inflation after 200 CE appears to have been far higher. Banks were the canaries in the Roman market economy, and they disappeared by the start of the third century. Argentarii had little reason to puzzle out the difference between real and nominal interest rates before 200; we infer that they were unable to do so fast enough after then to survive. Diocletian's Price Edict (Lewis and Reinhold, 1990, Vol. 2, pp. 422-26), one of several attempts to stem the inflation, reveals that many markets still were operating around 300 CE, but taxes in kind multiplied, and command economies grew. By the time of the Dark Ages, there were still markets, but no longer a market economy (McCormick, 2001). Roman agricultural technology and city planning were abandoned, education decreased, and long-distance trade in bulk commodities vanished. Roman law was forgotten in Europe for close to a millennium, and the Pax Romana ended with the early Roman Empire.

Click to expand...

The declining institutional quality mean't that life and property became less well protected, which mean't an increase in violence and thus of abnormal bone fractures among the inhabitants of the western world:



The picture of barbarians increasingly raiding and looting towns and villages, killing people randomly, describes pretty well what was increasingly happening in the western world from it's ancient zenith, around 100 AD, until the 8th century AD, when nearly all traces of the "Glory that was Greece and the grandeur that was Rome" had already vanished.

Sources:

(1) Hong et al, Greenland Ice Evidence of Hemispheric Lead Pollution Two Millennia Ago by Greek and Roman Civilizations, Science, Vol. 265 no. 5180 pp. 1841-1843, 1994
(2) Hong et al, A History of Ancient Copper Smelting Pollution During Roman and Medieval Times Recorded in Greenland Ice, Science, 1996
(3) Edited by Wilson & Bowman, Quantifying the Roman Economy, 2009
(4) Edited by Scheidel et al, The Cambridge Economic History of the Greco-Roman world, 2007
(5) Patterson, C. C., Silver Stocks and Losses in Ancient and Medieval Times, The Economic History Review, Vol. 25, No. 2, 1972
(6) Cleere, H., The Iron Industry in Roman Britain, 1981
(7) Sim & Ridge, Iron for the Eagles, 2002
(8) ORBIS
(9) ironwhe1.htm
(10) Maddison, The World Economy: A Millennial Perspective, 2001
(11) Temin P., The Economy of the Early Roman Empire, Journal of Economic Perspectives, 2006
(12) The Ancient Economy: Methods and Problems, 2007
(13) Perkins B.W., The Fall of Rome and the End of Civilization, 2007
(14) Jongman W., Gibbons was Right, Crises and the Roman Empire, pp 183-200, 2007
(15) Broadberry, Gupta, The Early Modern Great Divergence, 2006
(16) Takashima et al, Japan and the Great Divergence: 730 - 1870, 2011
(17) Amemiya, Economy and Economics in Ancient Greece, 2007
(18) Hadrill A. W., Houses and Society in Pompeii and Herculaneum, 1994

Louise C said:

The 'typical house' shown in his picture would only be 'typical' for the wealthy, the common people were mostly crammed into shoddily built apartment blocks.

Click to expand...

No, that was typical of Pompeii, a second rate Roman town. Also, not all apartment blocks were badly build, the ones that archaeologists have excavated were very well built.

I say again: the AVERAGE size of the Roman house was 2,800 square feet, at the first floor. Yes, the AVERAGE. The really rich ones had houses of 30,000 square feet, such as the house of the Faun:



This one was of above average size:


Here are a few Pompeii houses:


Notice that they vary greatly in size, however, the average is about the same as the image above, about 8 rooms, an atrium was typical of Roman houses. Most Roman houses had an atrium, not only the wealthy.

If you want to know more I suggest:

[ame="http://www.amazon.com/Houses-Society-Pompeii-Herculaneum-Wallace-Hadrill/dp/0691029091/ref=sr_1_1?ie=UTF8&qid=1340047372&sr=8-1&keywords=houses+and+society+in+pompeii+and+herculaneum"]Amazon.com: Houses and Society in Pompeii and Herculaneum (9780691029092): Andrew Wallace-Hadrill: Books@@AMEPARAM@@http://ecx.images-amazon.com/images/I/51GV1LR7ZqL.@@AMEPARAM@@51GV1LR7ZqL[/ame]

If you don't, I don't need to say anything else.

i_am_the_walrus said:

Well, this book: Amazon.com: The Fall of Rome: And the End of Civilization (9780192807281): Bryan Ward-Perkins: Books
is a very good and recognized read on this subject.
And if I remember rightly, its conlusions don't contradict Guaporense.

Click to expand...

No existing empirical evidence contradicts the fact that the fall of the classical civilization from the 1st to the 8th centuries represented a economic collapse of first order: People in the 8th century were simply much worse off than people in the 1st century.

The rest is (silly) ideology.

May I ask (excuse my relative limited knowledge on this subject) is there an equally recogniced alternative theory?

Click to expand...

No, there isn't.

Edgewaters said:

Pompeii was hardly "second rate" ... it was not a huge city if that's what you mean, but it was a very prosperous one ... it was a resort town for the urban elites. The "average" house there is hardly the average house in Rome itself. May as well go to Ibiza and say "this is how most Westerners live, most of the time"

Click to expand...

My sources such as this Houses and Society in Pompeii and Herculaneum Paperback – July 8, 1996
by Andrew Wallace-Hadrill, say Pompeii was a "third rate Italian town". Never heard of this conjecture that it was a resort town.

Mk Z said:

About copper production decline and iron production rise, could it be so iron just replaced bronze by the 2nd century for making weapons and tools and whatever?

As for lead, the graph says "exhaustion of Roman lead mines", so maybe it was shortage of lead/silver production which led to economical decline not otherwise?

And what do you think is the reason behind the economical collapse of the 2nd century?

Click to expand...

The long run economic collapse of the ancient civilization was caused by the dissolution of the City State. Ancient Civilization was based on the concept of individual citizen holding political rights because they were citizens of city states. This allowed citizens to hold property and to trade and stablish organizations like workshops and firms. After Rome conquered the ancient world they gradually centralized the Empire and crushed the relative autonomy of local governments. Eventually the empires population was reduced to serfdom and with it the economy collapsed since serfs have no incentive to be productive and and freedom to trade.

RoyalHill1987 said:

To the OP, great post.

A quick question - according to what I've read in bryan ward perkins' book, 'The fall of Rome and the end of civilisation', economic activity was actually still rising in the eastern Roman empire until the early 7th century, until the Persian wars followed by the Arab invasions ushered in the start of the dark ages.

Also, I understood that economic activity actually reached its nadir in western Europe around the years 700-800, after which a gradual recovery began that accelerated into a huge boom in the 12th century. That would suggest not a thousand years of decline, but more like a catastrophic collapse circa 600AD, followed by a recovery some 350 years later.

Any thoughts?

Click to expand...

Well, decline began around 150 AD and continued up to around 750 AD. I'm the Eastern Half the decline happened earlier in fact, since 150 BC I would think based on the volume of amphorae found with the stamp from Rhodes.

There was a relative recovery from 300 Ad up to 500 ad but even at that time it was a far cry from the peak much earlier.