The Prevalence of Talent
There are limits to what a person is capable of. In a given population, maximum skill lies along some sort of distribution. If everyone were to devote themselves totally to a single pursuit in a given society, people would eventually reach a point where there was a compressed distribution of that skill close to the maximum possible within that society.
To put this in concrete terms, if everyone in the US decided to devote themselves to football, we would eventually (within a generation or two) see every person within a couple percent of their maximum possible skill in football. The variance could be explained by factors such as lack of motivation, choosing a bad position (a person whose physiology was more suited to being a linebacker devoting themselves to becoming a quarterback, etc.), and normal imperfections of the world.
Ignoring other follow-on consequences of such an arrangement—such as the selection pressure to breed more capable children—a lot could be said about the population in a world like this. For instance, the total market for talent would be nearly saturated. Many of the people who are professionals today would not be in such a world because of selection pressures. Most people would be better at evaluating others' skill at the common task—not good necessarily, just better. Given that evaluation of skills and performance of them are correlated yet distinct, not everyone would be suited to determine which of their fellow citizens is better than others.
Consider: in a given domain, not only are there different specialties (quarterbacks and linebackers; percussion, strings, and wind), there are also different meta-specialties and synergies. A football team, due to the nature of competitive games, needs trainers, managers, coaches, strategists, data analysts, businessmen, and so forth. What makes a team successful in a competitive sense (winning games) and what makes a team successful in a business sense (bringing in as much money as possible) are somewhat distinct skills. Getting the attention of customers and getting them to want to spend money is in part due to the prospect of being excited—close scores, a memorable experience, novelty (being part of something no one else is), camaraderie (being part of something others are)—and not necessarily seeing one team utterly destroy another. A single star player is not enough to produce long-term interest: if no one can ever get the ball to the running back or the defense crumples under an offensive drive, no one is going to be interested in the "product" over time.
The football metaphor is limiting, so I will set it aside for now. Consider the following, then:
There is nothing that everyone does well, so we don't know what the limits of performance are. Given that time and attention are limiting factors in most business relationships, we tend to choose products and experiences not on raw quality but rather on some combination of the above factors, and others: exclusivity, inclusivity, identity, memorability, compatibility, reputation, exhaustion.
When it comes to occupations, there are two main poles at which work accumulates: that which can only be done by many people, and that which need only be done once. The former is represented by the term "labor": cleaning, cooking, building, advising, teaching. The latter is represented (mostly) by the term "creation": design, invention, research, architecture. That isn't to say that this second set of tasks only need be done once, ever; however, designing a particular (e.g.) car allows for the creation of a practically unlimited number of cars, whereas the construction of a single car doesn't allow thousands of people to drive.