We called the agents in our example of comparative advantage the Smith family and the Jones family. But those agents are themselves complex in several ways.
We thought of them, for instance, as farming families, each with cattle that could supply leather and sheep that could supply wool. We noticed that some of the leather was used for shoes, and some of it for other leather goods. We suggested that some of the families' cattle were in the form of milk-giving cows. We noticed the possibility that one family might be experimenting with cotton, and we explored what effect that might have on inter-family sock trading. We imagined that one family used a horse-drawn wagon for transport, and that the other family might use a river to float goods to their destination. To say that the families made shoes and socks was a natural enough shorthand, given that we were using shoes and socks to illustrate comparative advantage. But it was, after all, an abbreviation of a variety of skills that the families exercised, of various activities that were part of the families' economies. That was clear from some of the reallocations of labour we imagined, and we could have continued the list of possible reallocations. One family, for instance, might transport not only its own goods but some or all of the goods produced by the other family. One family might produce more or less fodder for the two families' livestock. As Xenophon and Adam Smith observed, a given economic activity can often be analyzed into component activities.
A given agency can also be analyzed into component agencies, and the competencies of the composite agency are typically not evenly divided among the components. The Jones family's talent for shoe-making, for instance, might derive entirely from Jim Jones's craftsmanship, and that would suggest various possibilities and limitations. The potential labour reallocation between the Smiths and the Joneses will require a labour reallocation within the Jones family. Is Jim willing, or even able, to take on the additional shoe-making? Who will now do the work that Jim gives up in order to concentrate on shoes, and how well will that work be done? Who, in turn, will take on the work given up by the member who takes on some of Jim's former duties, and how well will that work be done? How readily could Jim teach shoe-making skills to other family members? Is there room for a second cobbler's bench, without displacing some other useful piece of equipment? If the reallocations result in a staffing strain on the Jones's dairy operations, how vulnerable is the family in the event of a death among the dairy staff?
The same worries can be caused by external forces even when the labour allocation within an agency has remained steady for some time. Perhaps, after a time, the shoe-making business becomes less profitable; then the Joneses would have to allocate labour away from shoe-making. But perhaps by then, their sock-making operation has atrophied as a result of long-time concentration on shoe-making, so that the sock-making operation can't be sufficiently revived. On the other hand, perhaps the dairying operation's efficiency has increased, incidental to the allocation that supported shoe-making and that moved some family members out of dairying and others into it.
This sort of analysis can be applied to agencies of different sizes, types and situations, and the results suggest further complexities of different sorts. Suppose, for instance, that instead of two families we have a village of fifty families. There will be a great many putative cases of absolute and comparative advantage, and a great variety of possible labour allocations. When the Joneses now think not just of the Smiths but of all the other families as well, they will see new difficulties but also new safeguards.
Other families might be competitive in any of the Jones's constituent economic activities - dairying, woolen goods, shoe-making, and so on. The comparative shoe-making advantage the Joneses had with regard only to the Smiths might not obtain against other families or combinations of families. On the other hand, if the configuration does favour their shoe-making specialization, the larger market will provide a cushion that the smaller market could not. The Smiths' need for shoes might dwindle, but that wouldn't much change the aggregate village demand for shoes. At the same time, though, the existence of other shoe-making families imposes the prospect of competitive innovation - another family might introduce a different material for soles, or discover better ways to minimize waste in cutting shoe components from leather hides. The economic ecology of the village is different from the economic ecology of two families in a wilderness, and that influences the economic ecologies internal to the constituent families. With a good initial market for shoes and a reasonable opening advantage in shoe-making, the additional advantage obtained by specialization itself might make in-family allocation of labour to shoe-making more attractive than it was in the two-family situation; under different conditions, specializing in shoes might be less attractive.
Now our talk about families, villages, and their traditional trades has suggested something like a nineteenth-century rural setting for our imagined economies. It was only a few paragraphs ago that the cotton gin was on the verge of being invented. That setting tacitly carries with it some assumptions that should now be made explicit. We have probably imagined a moderate pace of technological change, and a relatively stable range of products in demand. There might be a sawmill in the village, and perhaps a grist mill, though probably not yet a textiles mill. In the main, products are made as they have been for some time, and the same products are wanted now as were wanted before. That, of course, isn't altogether untrue of any economy. I have been accumulating shares of Canadian banks and utilities most of my life, thinking as I do that the demands for electricity and for borrowing money will continue indefinitely into the future. I have avoided investing in very new things such as start-up "dot-coms" and legal cannabis producers, and that isn't so very different from a nineteenth-century investor not wanting to gamble on novelties - Mark Twain, for instance, ought not to have bet so heavily on the Paige typesetter. Still, in an economy of traditional trades and products, participants can rely on shoes and socks, milk and cheese, flour and harness, and indeed nearly all of the ordinary furniture of their lives, not to go out of demand at all or at least not quickly. Their village economy is very durable.
It is also an economy about which, on account of its size, the participants are quite well informed. The Joneses know what the Smiths do and what they might be capable of doing, and the same with the Johnsons, the Robinsons, the Harrisons, and so on. With minor variations, the same small group shares the same fall fair prizes in the same way, year after year, for tastiest pie, fanciest sewing, and best heifer. Children do what their parents did, and follow them eventually into the same cemetery. There are always some surprises - the agent of a distant equipment maker arrives in the village, for instance, and offers farm implements more cheaply than the village can produce them. Enough surprises in a short time can ruin a village, but that happens infrequently; small surprises can usually be accommodated by modest reallocations of labour. Everyone knows what the general economic conditions in the village are, and everyone believes they are quite stable.
Families in the village will, for the most part, aim to achieve survival and prosperity using strategies similar to what other families use. Suppose, for instance, that the Johnsons and the Robinsons come to a trade arrangement in which the Robinsons produce the cheese for both families while the Johnsons produce the iron implements, and that this makes both families better off. This will tend to influence the Smiths and the Joneses in two ways. They will be more inclined to specialize and trade in shoes and socks. At the same time, the efficiency of their manufacture of cheese and iron implements will decline relative to the village average, so that they will be inclined to work out their own agreements with the Johnsons and the Robinsons - who will themselves be motivated to trade for shoes and socks. Most families will, as it were, keep to the middle. At the current level of our example's development, they will undertake limited specialization, which is more profitable than no specialization at all and less risky than more extreme specialization. The levels of specialization will increase gradually, as in Adam Smith's example of the hunter who makes more and more of his fellows' bows and arrows until he is making them all. Absent outside influences, and other things being equal, levels of specialization will follow population growth, moving gradually toward the highly specialized economic ecology that Xenophon thought characteristic of cities.
The same considerations that apply among families in a village apply also among villages when there are several in the same economic geography. Specialization and trade begin slowly, and as they increase, whole villages come to specialize. A village will place a higher value on the specialty goods that it trades most profitably, and this will cause a reallocation of labour among the constituent parts of the village - families, we'll say for now. The same sorts of problem the Joneses faced in reallocating labour among family members will now be faced by the village. And so on, as villages federate into still larger units - townships, counties, provinces, countries. At each stage there will be larger markets, providing larger cushions but also larger threats. And at each stage there will be questions of how specialized to become. The size of casualties will become larger - as individual families in a village fail to find successful strategies, so individual villages likewise falter.
The types of choice faced by the proximate constituents - the members in a family, the families in a village, the villages in a county - thus tend to replicate at the next higher level. But as the total economy grows larger, the choices faced by more remote constituents of that economy do not remain as they were. What counted as "keeping to the middle" for a family in an economy that ended at the village level will change as the village begins to function in a larger economy. Conceptually the family's strategy for survival and prosperity remains the same - keep to the middle. But keeping to the middle in a village that itself must now specialize among other villages will mean that the family must now tailor its own strategy to the overall strategy of the village. And the same with individual family members. The choices open to more remote constituents of an economy - the atoms rather than the molecules, we might say - are more constrained that the choices open to more proximate constituents.
What happens to constituents who aren't suited by these narrower specialization constraints? One remedy is relocation. A family that really doesn't like the activity in its village can move to another village. Family members who don't like their family's activities can - when they become adults - relocate to another family by leaving home. But larger units have fewer relocation opportunities. A village, province or country might, over a longer period, begin to change its trading partners, thereby "relocating" itself into a somewhat different economy. But no literal relocation is possible, because larger entities have inherently geographical characteristics. The only way really to move a village is for its constituents - in our example, the families - themselves to move.
A second buffer against unwanted specialization is the facilitating services that invariably spring up as specialization increases. Jim Jones might not be happy doing nothing but making socks, for the same reason that he would be unhappy doing nothing but making shoes. But the shoes and socks, the cheese and the iron implements, will need to be delivered - and counted, inspected, packaged, stored, marketed, and so on. And it is often the case that, although someone who makes shoes does nothing but make shoes, someone who delivers shoes also delivers socks, and perhaps arranges for their storage. There is variety to be found in the spaces between specialized production. And as the services themselves become specialized, there will be variety in the spaces between specialized services. Moreover, the sheer increase in the number of specializations makes it more probable that a given agent will find a congenial specialization without resorting to the spaces between specializations.
Two broad claims have now emerged in our analysis. (1) Economic advantage, whether absolute or comparative, suggests specialize-and-trade strategies; but it is extremely difficult to determine whether an apparent advantage retains that character when all things are considered. Immediate opportunity costs almost certainly do not reflect comprehensive opportunity costs. (2) Despite that difficulty, economies do tend to rely more and more on specialization and trade as they get larger, as Xenophon and Adam Smith thought they did. Together, the two claims suggest that larger economies must involve a great many mistakes about what is really beneficial. Now the availability of specialization alternatives is a natural economic mechanism that cushions the effects of those mistakes on the various economic agents. And additional mechanisms such as social insurance and welfare can be put in place deliberately to provide further cushioning. But whether the natural and the artificial cushions are globally efficacious, or whether they can remain so, must seem tentative and conditional. The remainder of this essay is devoted to exploring that question.
Comments