A horse ! A horse ! My serfdom for a horse !

Part 2, England, of my critique of Nick Szabo’s view of industrialisation. This is continued from Part 1, “Chinese workers were cheaper than English horses“.

What do coal, American slave cotton, and the resources of the New World have in common ? All of them were important in the English industrial revolution, but none was crucial to getting it started and none might have been absolutely necessary even in the long run. All of them had second-best substitutes which would have been more costly and therefore lowered output. Their absence would have slowed the pace of industrialisation, and changed England’s pattern of trade and specialisation. But industrialisation would not have been stopped dead in its tracks. And the pace would have been slower only until the coming of other innovations, especially those technologies associated with the “Second Industrial Revolution“.

Horses are not even in the same league as coal and cotton, despite Nick Szabo’s argument in his blogpost, “Transportation, divergence, and the industrial revolution“.

I have absolutely no problem with the conventional idea that transport networks are really really important. I just have a problem with the idea that a technology which diffused in 1200-1300 (the draught horse) with a high degree of substitutability (compared with, say, coal and charcoal) should have such a massive importance in 1800.

Industry Location

Industry location was the most important way of dealing with transport costs. In the earliest phase of English industrialisation (1750-1800), industry moved to the sources of its most expensive inputs. Iron works were located near all three ire ore deposits, collieries and canals/ports. The textile mills were also located basically atop coal deposits and water sources. But until the 1830s coal-powered steam was not widespread in Lancashire’s cotton textile industry anyway, which had relied mostly on water power. See below for clickable thumbnail maps of English industrial concentrations in 1700 and 1800 (source). I wish I had a similar map for 1600, which would show a strong clustering of industry in southeast England.

cameron1 cameron2

Transporting coal

Szabo says, “The early industrial revolution was highly dependent on bringing together bulk goods such as coal and iron ore”.

Because the cost per mile of water transport was so much smaller than the costs of land transport, this “last few miles to the mine” problem usually played a dominant role in transportation economics, somewhat analogous to the “last mile” problem in modern cable networks.

The “last few miles to the mine” issue apparently played little role in the real-world economics of English coal in the 18th century. From Clark & Jacks 2007 :

Extraction costs, rather than transport costs, seem to dominate in the determination of rent [value of output minus total cost]. If transport costs to water were a substantial cost relative to rents then we would expect to see that after the introduction of rail travel c. 1830 there would be substantially greater dispersion of the pits from the Tyne. In fact average distances changed little. We calculated how close each pit was to the water, which varied from 0.25 miles to 18 miles. In 1727 to 1829 the average distance to the Tyne was 5.4 miles, in 1831-1864 6.4 miles. But this difference is not even close to statistically significant. Presumably the existing network of wagon ways in the Tyne area was a fairly cost effective way of getting coal down to the Tyne. Second if transport costs to the Tyne were substantial we would expect to see lower rent at more distant pits, since there transport costs would absorb all the site rents. There was, however, no correlation between rent and distance to water, even before the arrival of the railroad. The implication of this is that the transport costs to the water were generally low, so that they did not have much impact on mineral rents. Thus the site rents measure mainly differences in extraction costs.

In fact, the transport-cost share of the real price of Newcastle coal delivered to London barely changed until the 1820s !


The major drop in transport costs after 1820 may reflect the earliest use of steamship for coastal shipping. Yet, well before the 1820s, there had been quite dramatic reductions in the cost of overland transport in the rest of the British economy. From Mokyr :

transport productivity

In the particular case of coal — the quintessential high-weight, low-price resource of the Industrial Revolution — there is no evidence for Szabo’s contention that “the early industrial revolution…was very sensitive to small changes in land transportation costs”. Apparently location took care of the issue. So imagine what the cost structure would be like for goods with higher value-to-bulk ratios, like textiles and iron, which were not only located near their sources of power, but also located near ports, rivers and canals.

English transport costs already cheap in the Middle Ages !

Overland transport was already pretty cheap in England by 1300. In “Transport Costs in Medieval England” James Masschaele compares overland transport prices between the 1350s and the 1750s. When adjusted with a consumer price index he finds that costs in ~1350 were 40% lower than in ~1750. But just in case that index is not trustworthy, he converts the costs in each period to wheat prices. The results are not substantially different. This can mean, either that overland transport was really more expensive in 1750 than in 1350, or that wheat was much cheaper in 1750. The latter would imply nonetheless that transport productivity hadn’t improved all that much.

The low cost of mediaeval transport in England is confirmed by other data obtained from a different method : the regional integration of the grain market in England over the same time period.


[Source] In a well-functioning market, the difference in prices of the same good across regions should reflect transaction costs, including costs of transport. The gap was low in England quite early. You do get even more improvement after 1700, which you would expect, given the transport productivity growth outlined above. It’s possible transport productivity growth had a much bigger impact on distribution to consumers rather than the sourcing of inputs.

Transport costs are not magical

“…we reach a dramatic but solid mathematical conclusion: the potential value of a land transportation network is the inverse fourth power of the cost of that transportation. A reduction in transportation costs in a trade network by a factor of two increases the potential value of that network by a factor of sixteen. While a power of exactly 4.0 will usually be too high, due to redundancies, this does show how the cost of transportation can have a radical nonlinear impact on the value of the trade networks it enables.”

You can do a “social savings” calculation for the impact of the improvements to overland transport on the total economy. The growth in output in excess of growth in inputs (total factor productivity) in the British overland transport sector was 1.59% per year between 1700 and 1860. That’s a lot. It amounts to a 12-fold increase during those 160 years. The share of the overland transport sector in British national income was about 6%. TFP growth in the overall economy was about 0.56% per annum over the same period. (It was lower in 1760-1800, higher in 1800-60.) So savings in the transport sector account for about 1/5 of the growth in the total economy in this period. (And, yes, I’m aware of externalities.) [Source for these figures is the latest edition of The Cambridge Economic History of Modern Britain, Volume 1, chapter 13.]

But that’s the economic impact of the improvements to overland transport, not the impact of Szabo’s hobbyhorse. I suppose some enterprising drudge could duplicate for England in 1800 what Robert Fogel did for American railroads in 1890 : assess their impact on the US economy by calculating the “difference between the actual cost of shipping goods in that year and the [counterfactual] cost of shipping exactly the same bundle of goods between exactly the same points without the railroad” but with more canal development. (Fogel 1979) Similar calculations have been done for railroads all over the world (but especially British), steam power, computers, telecommunications, and other “general purpose technologies”. Such a simulation would be much easier for horses.

The effects are always non-trivial but more modest than one would expect from the glamourous repute of the technologies, either because the innovations improve welfare more than add to output (e.g., passengers spend less time on travel ; the iPhone doesn’t necessarily improve my productivity, may worsen it, but improves welfare) ; or because the second-best substitutes are never that bad (contrary to Szabo’s Law of No Second Bests) ; or because the large effects are lagged and smoothed over long periods.

About pseudoerasmus

This entry was posted in great divergence, Industrial Revolution and tagged , , , . Bookmark the permalink.

3 Responses to A horse ! A horse ! My serfdom for a horse !

  1. Pingback: Chinese workers were cheaper than English horses | Pseudoerasmus

  2. Erik says:

    Links to part 1 are broken.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.