Robinson dismisses Smith by suggesting that people can equally divide their own labour across different tasks through time. The 18 distinct operations Smith recounts could just as easily be conducted by the same labourer on 18 different days to generate the same output per person over an 18 day period as in the case where labour is divided between workers.
Further, the fact that relatively unskilled labour could perform any of these tasks adds to the case that it is not the division of labour at play in generating productivity gains. One-way causality from the division of labour to productivity gains is a highly problematic story.
But that leaves open the question about the actual mechanism that provided the enormous productivity gains in the pin factories of the mid-1700s.
Instead of Smith’s division of labour hypothesis, let me propose a capital investment hypothesis to explain the productivity of his pin factory. This hypothesis suggests that it is the technical nature of capital that determines the way labour will be divided across tasks to maximise output and that the division of labour responds to capital investment. The causality goes from capital investment to labour division.
To guide my inquiry I use the structured approach I have advocated for in the past, confronting economic issues by first asking questions about aggregation. For example, why are there 18 tasks to make a pin? Why are 18 workers in one pin factory and not 9 in one factory and 9 in another owned by a different entity?
The answer to these questions is capital. The image above (source) shows the tools and equipment used in the pin factories described by Smith. Notice that the tools and machines in the picture have been designed to more efficiently perform distinct parts of the pin-making process. It is the way the tools have been designed to efficiently break down the task of making pins that leads to the labour division to ‘man the tools’.
Smith came close to instead presenting the capital investment hypothesis. He says
…a workman not educated to this business (which the division of labour has rendered a distinct trade), nor acquainted with the use of the machinery employed in it (to the invention of which the same division of labour has probably given occasion), could scarce, perhaps, with his utmost industry, make one pin in a day, and certainly could not make twenty. [my emphasis]He suggests that it is the division of labour that has probably given rise to the machines, rather than the machines themselves giving rise to the division of labour. But this logic comes undone later in the paragraph, even though he ignores the inconsistency in his argument.
…the important business of making a pin is, in this manner, divided into about eighteen distinct operations, which, in some manufactories, are all performed by distinct hands, though in others the same man will sometimes perform two or three of them. I have seen a small manufactory of this kind where ten men only were employed, and where some of them consequently performed two or three distinct operations. [my emphasis]Even based on Smith’s own observations it is the tools and machines that generate the 18 tasks since people can perform more than one of them. So how exactly how did the division of labour give rise to the invention of the necessary machines that generate 18 tasks with only ten men?
If it was the division of labour that led to increased productivity, labour could just as easily be divided between firms. The fact that pin factories, even with only ten men, still performed all 18 tasks, instead of specialising in just 10 tasks, is clear evidence that there is something special and coordinated about the tasks themselves that arise from the particular capital investments. The tools and machines are designed to be compatible with each other, and if part of the process is done outside the firm, each of the two firms would inevitably be tied to the same compatible capital equipment, and would therefore find gains by merging into a single firm.
When we ask questions about the timing of investment in machines we can more sharply distinguish between the division of labour and capital hypotheses. If it was only after the machines were introduced that labour was divided in a particular way, then that is evidence for the capital hypothesis. If labour was divided into 18 tasks prior to the investment in machines, achieving the same tasks in the absence of tools, then the division of labour hypothesis holds.
Indeed, the prediction of the capital investment hypothesis is that labour task specialisation responds to capital investments in either direction - either with the division of labour, or consolidation of tasks by a single labourer.
A modern test of these predictions could be garbage collection. With rear-loading trucks, labour is divided between driving the truck and loading the bins. But with more advanced side-loading trucks with robotic arms, the labour is once again undivided between driving the truck and collecting the bins. The progression of capital technology determines the division of tasks.
Like many stories in economics, the division of labour as a productivity-enhancer has been approached far too narrowly. There are many economic lessons in the story of the pin factory, and if we probed deeper we could understand more about what considerations determine the boundaries of firms, and why firms are internally not-structured around market principles. But a structured approach to economic inquiry is required to turn Smith’s pin factory into a useful learning tool. The same applies to other stories, like economists favourite Robinson Crusoe story about specialisation and equilibrium. But that will have to wait for another time.
Perhaps it is unfair to make a judgment on Joan Robinson based on the account presented here, but she must have been totally out to lunch by focusing on Adam Smith's anecdotal observation about a pin factory. The broad division of labor that we have in modern, civilized societies is a result of the application of science, engineering, constitutional governance and the freedom of individuals to do what they want with their lives. Our society has become increasingly complex with the proliferation of new products and services. It is only the division of labor that makes this possible. We can exchange our abstract expertise in executing a task, be paid in currency that can be exchanged for other goods and services that rely on the division of labor. Even Piketty was honest enough to observe that even though inequality has increased to the point where it matches that of the ancien regime, the standard of living has increased at least ten-fold since that time, rendering today's poor better off in a material sense than even a successful lawyer in the age of La Pere Goriot.
ReplyDeleteThe pin factory observation is repeated religiously and without insight in every single intro econ textbook, so it's right to question it.
DeleteCameron K Murray is both right and wrong. For example:
ReplyDelete“The 18 distinct operations Smith recounts could just as easily be conducted by the same labourer on 18 different days to generate the same output per person over an 18 day period as in the case where labour is divided between workers.”
The same labourer undertaking the 18 operations on different days would take 18 days to produce the completed pins for shipping to customers. Whereas 18 labourers using the machines as illustrated in the original French Encyclopedia from which Smith took the pin case (see Paucelle, J. L. 2006 “Adam Smith’s use of the multiple references for his pin making examples”. European Journal ofthe History of Economic Thought, 13(4) pp. 480-512), were all using machines (capital) for the separate operations and produced their total 18,000 output on a production line every day or so. This output was available for shipping to markets almost daily and not just ever 18 days.
The drawings we have of pin factories shows several machines at work attended by individual labourers. Of course capital was required to manufacture the output of pins! The pins were cut by metal cutting hand operated machines (like guillotines), and the heads soldered by labourers at machines. Similarly with placing the pins in paper packets.
Over the decades multiple pin plants grew up across the country as machine-driven power became available in the 19th century,with a few considerable-sized centres specialising in pin manufacture appearing. Then in the 20th century larger plants appeared producing most pins in even fewer companies. With automation in the later 20th century, a handful of companies produced the country’s pins, until recently there were one of two plants produced all the UK’s and the USA’s pins in multiple-production lines (without 18 labours!).
However, since the early 18th century the division of labour was always accompanied with capital investment. Smith was aware of that clearly (such illustrations featured throughout the encyclopedias). Capital enables the productivity of labour to grow and the division of labour facilitates that labour productivity growth.
"The same labourer undertaking the 18 operations on different days would take 18 days to produce the completed pins for shipping to customers. Whereas 18 labourers using the machines as illustrated in the original French Encyclopedia from which Smith took the pin case (see Paucelle, J. L. 2006 “Adam Smith’s use of the multiple references for his pin making examples”. European Journal ofthe History of Economic Thought, 13(4) pp. 480-512), were all using machines (capital) for the separate operations and produced their total 18,000 output on a production line every day or so. This output was available for shipping to markets almost daily and not just ever 18 days."
DeleteIf each of the 18 individual workers started their 18 day production cycle on consecutive days, then the aggregate output of the divided and non-divided labour situations would be the same. Each day's output would be identical.
It's good to see that someone else noticed this.
ReplyDeleteI always thought the issue was that tools were expensive, so one worker could not afford to own 18 different tools and leave 17 of them idle at any given time. Collective action with enough workers to keep all the tools active would be more efficient. The big debate has always been about the nature of this collectivization.
This suggests a countervailing force as manufacturing equipment decreases in price and increases its general capabilty. If everyone has a 3D printer that can work with a variety of materials and costs as much as a microwave oven, then we could move to a more individualist model.
I totally agree with your first paragraph, which I hope is a point that come across in the post.
DeleteBut I don't think there is any unique property of 3D printing compared to other manufacturing processes that would change the nature of organisation around capital. After all, between each print using your 3D printer will be a design period where the printer will be idle. There will also be economies of scale from larger printers. So the same dynamic is at play in general, even if it does open up more possibilities for sophisticated small-scale production organisations.