The Limits of Individual and Intergenerational Specialization

Chapter 11 of The Information Class

As a grim global breakdown unfolds in the wake of Trump II and Tariffmageddon, I’m publishing chapter-by-chapter a mainly theoretical book about class that was mostly fully written in 2024 and earlier. It seems an odd and even escapist pastime, and I admire those like Richard Hanania and Brad DeLong who are manning the intellectual front lines of fight for democracy, capitalism, the rule of law, and a scientific civilization. But what I call “the information class”— college educated people who work with their minds using computers— is still here, staffing most of the organizations and writing and sometimes organizing, and loosely standing ready to restore competence and decency when nihilistic populism burns itself out. In these posts, I’m trying to shed light on them, help them know themselves and each other, and foster the integrity, solidarity and noblesse oblige that they’ll need to rise to the challenge of these dark but not yet disastrous times.

To get up to speed, read the intro, chapter 1, chapter 2, chapter 3, chapter 4, chapter 5, chapter 6, chapter 7, chapter 8, chapter 9, and chapter 10. And sign up to get future chapters in your inbox.

How far should individual specialization and occupational differentiation go? And how far can intergenerational specialization go? How far should it go? How many classes should there be? We've grasped the meaning of a class in the context of a larger economy of occupational differentiation, but what does the overall social structure tend to look like?

There are no certain or final answers to these questions. Societies differ in the class differentiation that is practiced, as well as the class differentiation that is optimal, given the human capital endowments of the population and the state of technology.

And yet much can be said, all the same. There are some generalizations that can be offered.

Economists are less insightful about this than they should be, because they too often take the lazy dodge of “leaving it to the market,” and thereby evade many difficulties of planning. That's a mistake. Of course, it would be foolish to try to micromanage everything about how broad or narrow people's job descriptions are. But government, to which economists are often called on to serve as advisors, has a large footprint in the economy, and influences patterns of specialization in countless ways, from educational curricula and student loans to occupational licensing to labor law to contracts enforcement. Government can't really stand aside from the question. Since it has to play a role, some attempt at intentional planning would be appropriate, and economists and pundits should do their part by floating specific ideas, rather than just “leave it to the market.”

As we've seen, specialization is incredibly valuable. More than any other single factor, an effective system of specialization and trade is crucial for high productivity, a strong economy, and high standards of living. And I don't think there is any particular inherent limit on how far specialization can keep getting elaborated and improving economic performance. In general, whenever a person does a job, then if there were ten of them and they could focus on different aspects of it and trade, they'd be more than ten times as productive, since they could be more focused and invest in more task-specific skills and tools. There’s no reason to foresee any upper bound to the applicability of this principle.

But in other senses, there are limits.

First, there is the simple limit of population. You can't have more specialists than you have people. The relevant population will differ by the reach of logistics. Haircuts are a local service, because a barber in India can't cut your hair in New York. For most widgets, the relevant population is global because they can be shipped long distances. So workers and machines in widget factories are a lot more specialized than barbers.

Second, there is a factor that doesn't have a good name, and is a bit hard to explain. It's a kind of functional connectivity that is sometimes called “social capital,” and relates to the structure of social networks, and to habits of trust. To understand it, start from this: I need x done; you can do x; and your reservation wage for doing x is lower than my dollar value for having x done. The circumstances for mutually beneficial trade in x are present. But do I know you exist? Do I have a way of finding you? Can you and I make a deal and trust each other to hold up our end? Such awareness, searchability, and trust comprise the kind of functional connectivity I have in mind. In its absence, there is specialization failure.

Third, some limits to specialization are set by the fact that to be unique is to be doubly vulnerable: vulnerable if economic flux eliminates your narrow function; and vulnerable in your bargaining position vis-a-vis counterparties who have more options.

In this connection, there's a funny saying in English: “Don't put all your eggs in one basket.” The metaphor shows its own limitations if you think about a shopper in the marketplace deliberately carrying multiple baskets with a few eggs in each. Not only does it require more effort, but if anything, you might be more likely to break all your eggs through the clumsiness of holding multiple baskets, than if you just took the one basket and handled it very carefully. And that principle also applies to careers. If you try to run two or three careers in parallel, you're likely to fail at them all.

Still, it is usually wise not to let yourself get pigeonholed into too narrow of a function. You want to try to diversify your skills, tasks, and customers somewhat.

It may be helpful to think of a “specialization frontier,” in which the economy is comprehensively characterized by the maximum specialization that population and connectivity permit; and then to say that the optimum lies somewhat behind that frontier. At the frontier, there would be too much rigidity, too much wasteful tug-of-war style bargaining, and too much systemic fragility. A little ways behind the specialization frontier, there is redundancy, resiliency, optionality, and competition. That's what people choose, and that's what's best for society, although I wouldn't want to suggest that self-interest will motivate the socially optimal amount of specialization, either. I see no reason to think the limits of specialization due to the social benefits of redundancy and resiliency precisely match the limits due to individuals’ defensive diversification. Policy might sometimes, in different situations, improve the economy in either direction: by nudging people to expand their range of skills, or by nudging people to focus and pursue specialized excellence.

Personal specialization is one thing, class specialization another. As we've seen, it's an advantage for productive specialization if you can learn a family trade starting in early childhood. But buy-in for that from both parents is likely to be forthcoming only if both of them have a background in that occupation themselves. Pure intergenerational occupational specialization requires occupational endogamy. But an occupation typically won't provide marriage market critical mass. There probably won't be enough candidate spouses among the offspring of people in your parents’ occupation. Even if there's a decent number of such people, their children might have an imbalance of boys and girls, or the wrong ages, or simply a lack of physical attraction. You want to have a wider selection than that. And so classes are less specialized than people are, pooling several or many occupations into one class.

Here the Indian caste system is a kind of exception that proves the rule. As far as I understand, many jati, or traditional Indian occupational castes, long practiced occupational endogamy, and it gave Indian civilization great stability, and great rigidity. It's better to pool occupations into classes than to try to make them naturally endogamous.

Class can be thought of as a kind of midpoint on a spectrum, against the extreme of equal opportunity classlessness, on the one hand, and of hereditary occupational caste, on the other. Class lies upstream of occupation on the specialization tree, and so it provides a larger marriageable pool, and naturally gives rise to marriages able to endow offspring with both a headstart and substantial optionality. Class, at its best, fosters a society that is abundant in the pursuit of specialized excellence, and has substantial resources of community and cohesion, but which is also adaptable and creative.

So how many classes should there be? Not as many as the number of occupations. It seems like most class-stratified societies have three or four at most. As with occupational differentiation, while government can't completely stand aside, much must be left to the market. In the end, a healthy class is a consensual phenomenon. Its members want to be part of it, and find it in their interest to maintain it, while others willingly treat members of the class differently, finding it in their interests to do so because the class has certain characteristics that make it wealthier, more trustworthy, or whatever. Laissez-faire, with robust freedom of association, will naturally produce some class stratification.

There’s no reason to think that classes are perfectly honed by competitive or evolutionary forces to precisely meet the needs of a social role. Classes are certainly not in perfect competition with each other. They're far too different for that. Individuals within a class face competitive pressures from each other, which often push them in the direction of conformism, but the same conformism that constrains individuals is a kind of liberty for the class. If people conform to the fashion, the fashion itself is free, and in the arbitrariness and transience of fashion lies a remarkable testimony to the collective freedom that a class enjoys.

Class is easily mocked. Fashion is easily mocked. And fashions are often silly. But they can also be beautiful or benevolent. And classes have the freedom to do beautiful and benevolent things, because a certain internal solidarity combined with a certain lack of external competition gives them a lot of freedom, a lot of room for maneuver. This room for maneuver can be wasted in frivolous luxury, and scorn for the vagaries of mere fashion is a useful check on that. But the same room for maneuver can also be used for the pursuit of excellence beyond what is demanded by any immediate necessity.

In particular, the freedom of classes to adopt traits not directly adapted to their economic function gives room for noblesse oblige to flourish. Privileged people can be peer pressured into being generous, charitable, and gracious towards the less fortunate. Their advantages give the means, and social expectations supply much of the motive.

Indeed, a large part of civilizational progress arises from a synergy between the way individuals and classes pursue excellence and virtue. Classes practice spontaneous and pervasive “peer review,” of a kind more authentic and often more astute than the formalized and often corrupt peer review utilized by academic journals and tenure committee to try to measure, reward and incentivize forms of merit that they don't understand and lack the confidence to judge. Healthy classes consist of people united in pursuit of an ideal by which they are constantly judging each other, seeking to filter out and marginalize the unworthy, while recognizing, honoring and emulating the most deserving actions, careers and individuals, for the betterment of the class and the more effective pursuit of excellence.

And of course, all this peer review feeds into mate selection. You'd better have a good reputation among your class if you want to marry well. “Gossip” and “snobbery” are common in most civilizations and have an important role to play in maintaining them, even though the terms sometimes deserve their pejorative connotations. But sometimes they don't. The sharing of information about the virtues and vices of others, and the sorting of people into more esteemed and useful and less so, can be done in a prudent and Christian spirit that conduces to the common good.