KELLY Morgan (1997) “The Dynamics of Smithian Growth”, The Quarterly Journal of Economics, 112/3, 939-964.
Smithian progress (i.e. the understanding that specialization causes output to rise first articulated by Adam Smith) is neglected by economists because it is perceived as gradual (unlike the sudden growth caused by innovation, learning by doing and private capital accumulation). But the author dismisses the idea that Smithian growth is necessarily gradual.
A market geographical size which expands gradually will lead to “vertical disintegration” in production (i.e. specialization). The author intends to show that this gradual expansion does not create a gradual growth as previously assumed, but a sudden one after a certain density of market linkage (i.e. routes) is reached.
When the critical density is met, small local that limit the scope of specialization markets start to coalesce into a large market that spans the economy (939). In other words, it is a sort of Smithian take-off that reallocates labour from inefficient subsistence production to more productive market-oriented tasks. The author points out that specialization is not a self-sustaining engine for growth and that it will need to be relayed by others.
To test his hypothesis, Kelly will observe the role of market expansion in the economic transformation of Sung China. This take-off was allowed by the opening of a national waterway network. A reconstruction of the Chinese growth (proxy for productivity) rate shows indeed a sudden increase around the period when the network was open – i.e.10th century (942).
The author uses a model where several villages start by being autarkic but may invest and become commercially connected with each other (943). There are two types of producers, the traditional able to produce several goods and the specialized only able to produce one at a faster rate. The price of a good is determined by the amount of work put in it. Supposedly, as market expends prices fall but wages stay the same and this happens as soon as the liaison is open between two villages.
The larger the market, the more efficient the labour (as well as other factors) allocation, output rises and prices fall. Logically, if economies of scales have increasing returns (i.e. the bigger the market added the faster the output growth), the first investments in new linkages have a limited efficiency (connecting two villages together), but after a while the “threshold effect” kicks in and output growth increases sharply. But after that outburst the possibilities of growth are exhausted and for it to go on it requires new triggers (capital, innovation, learning by doing).
The integration of transport costs may delay the commercial revolution for the most costly goods to transports, but the less costly ones will know their “threshold” earlier (950).
Testing the theory in Sung China
Eleventh-century China saw a rise in the use of money, the size of plants and estates, the size of the market, the number of technology available (gun powder), etc. Demographically, it was marked by the shift Southward of the population fron the wheat-millet regions around the Yellow River to the rice region of the Yangtze basin. As peasants looking for new land and fleeing the nomad raids moved in, the South’s share of the population jumped from 25% in 600 to 71% around 1200. To carry fiscal grain to the North, the government had to extend the waterway network. By the end of the 10th century, all economically important regions were linked by a 30,000 miles long network.
Thus entire communities, suddenly in contact with each other, could swift from subsistence to commercial agriculture. But a 250 years gradual growth does not qualifies as sudden growth predicted by the article. The mass of money in use seems to have increased by 0.65% a year over the 10th and 11th centuries while as proof of growth output in the iron production increased 4 times over. But this 11th century peak was followed by centuries of steady decline. One of the reasons of this stop was high government taxation (13% of national income, against 5 under the Ming).
The author proposes to analyse other events under the light of his model: the 13th century “commercial revolution” in Europe. He compares the large trade network backed by strong states allowed, in Italy, by a high demographic density, while weaker states such as the Hanseatic League could not guarantee security over the roads of sparsely populated eastern Europe and had to rely on more primitive networks.