Perhaps I will later write an even longer and more ponderous post on Tyler’s take on comparative advantage . But for now I content myself to say that, while I agree with much of what Tyler says, I believe that the thrust of Tyler’s post misses the essence of the principle. That principle is – and has always seemed to me to be – one about the proper way to reckon costs. To understand comparative advantage is to understand that costs, properly reckoned, are always opportunity costs. To understand comparative advantage  is to appreciate the reality that comparing how many baskets of broccoli Smith can produce in an hour or in a day or in a year compared to how many baskets of broccoli Jones can produce in the same amount of time does not tell us whether the cost of employing Smith to produce broccoli is higher or lower than is the cost of employing Jones to produce broccoli. Another comparison is germane – namely, the (subjective) value of what Smith gives up to produce a basket of broccoli compared to the (subjective) value of what Jones gives up to produce a basket of broccoli.
It’s a misfortune that the principle of comparative advantage was introduced to the world in the context of an explanation of international trade . Nothing about that principle is unique to specialization that emerges in response to expanded opportunities for individuals to trade internationally. Generations of careless scholars – including generations of careless economists – have been mislead into supposing that today’s existing matrix of comparative advantage serves either the exclusive – or a hyper-disproportionately important – role in determining patterns of international trade. In fact, today’s existing matrix of comparative advantage is
(1) not static; it changes over time;
(2) only one of the ‘forces’ working to determine, even in laissez-faire economies, prevailing patterns of specialization and trade;
(3) no more likely to produce ‘better’ patterns of specialization and trade than are other forces at work in free markets to determine patterns of specialization and trade; and
(4) always at work influencing patterns of specialization and trade intranationally no less and no more than it is at work internationally.
Of course, comparative advantage can be, and has been, explained to work not only ‘statically’ – that is, with a matrix of given and fixed opportunity costs for all producers – but also dynamically. The latter might explain, for example, how Smith, who today has a comparative disadvantage relative to Jones at producing broccoli, gains a comparative advantage over Jones tomorrow at producing broccoli because Smith has a comparative advantage today and tomorrow over Jones at innovating – or at investing, or at ‘waiting,’ or at experimenting – in broccoli (or agricultural) production. My favorite example: when I was 18 years old I certainly did not have a comparative advantage at teaching principles of microeconomics; now, as I approach my 55th birthday, I do have a comparative advantage at that task. The reason, of course, is that I set out, starting when I was 18, to eventually gain such a comparative advantage. The fact that I once bought instruction in economics principles but today produce and sell such instruction in no way violates the principle of comparative advantage. All along, even when I was 18, I likely had a comparative advantage at preparing myself to gain a comparative advantage at teaching economics principles.