A machine becomes productive in the hands of people who know what to do with it. The knowledge may be visible in a design office or almost invisible in the adjustments of an experienced worker: a sound heard before a bearing fails, a measurement taken before an error spreads, a better sequence of operations learned over years. Human capital gives an economic name to acquired capacities embodied in people. Education, training and health can require resources now and enlarge what a person can accomplish later. Gary Becker's account places these investments beside other forms of capital while emphasizing their embodiment: the knowledge travels with the person who possesses it. The distinction matters to the question this plate asks. A nation's machines and its capabilities work together, and counting the machines alone leaves out much of what makes their use possible.1
Acquiring a capability costs time as well as money, and the person who can use it must be able to reach it. A training course takes an instructor, equipment and hours in which the learner might otherwise have earned wages. Practice can consume materials and interrupt production before it improves it. The eventual gain may be large while the immediate bill remains impossible to pay. A person cannot ordinarily sell a spare portion of next year's skill to cover this month's rent, and a lender cannot repossess completed learning as it repossesses equipment. Employers face a related calculation when another firm could hire the trained worker. These problems help explain why the existence of a valuable skill does not ensure that everyone who could acquire it will do so. The return, the means of financing the cost, and the opportunity to use the result belong to the same investment decision.
A small calculation shows what must be compared before calling training worthwhile. Suppose a course costs twenty today, including the earnings sacrificed while attending it, and produces an additional six at the end of each of the next four years. At a five-percent discount rate, those four payments have a present value of about twenty-one and a quarter, leaving a net gain of about one and a quarter. If the additional earnings last only three years, their present value is about sixteen and a third, and the investment no longer covers its cost. Duration matters alongside the size of the annual premium. So do uncertainty, employment opportunities and the earnings the person would otherwise have received. The calculation concerns the learner's monetary return under the stated assumptions. It has yet to establish whether the extra wages represent extra production, a changed division of existing production, or a payment for information about the worker.2
Education can raise earnings by changing what an employer knows as well as what a worker can produce. Michael Spence's signaling model shows how an educational choice can convey information when employers cannot directly observe applicants' productive capacities. In his example, acquiring the signal is less costly for the more productive group, so an appropriate wage schedule can make the groups choose differently even when education itself adds no productivity. That possibility changes the interpretation of a credential's reward. It also requires care: Spence recognizes that better information can improve the allocation of people to jobs, so a signal need not have no social value merely because it leaves individual capability unchanged. Learning and signaling can coexist. The practical question is how much of a particular expenditure builds a capacity, how much helps others recognize it, and whether the same useful information could be obtained with fewer resources.3
A capability can also benefit people beyond the person who acquires it. A worker may show a colleague how to solve a problem, or make a team better able to recognize and use an unfamiliar technique. Robert Lucas gives this possibility a place in growth theory by allowing average human capital to affect productivity in addition to the return a person receives from individual skill. When that external effect is present, private decisions leave some benefits out of account. His construction is a model of how such a difference could matter, with sustained growth depending on the specified process of accumulating human capital; it is not a measurement of an automatic dividend from every additional diploma. The distinction also protects the meaning of a public good. A technique can be communicated and used by others, while a skilled person's working time remains limited. The lesson that spreads and the person who teaches it are different economic objects.4
Measuring the stock of human capital requires deciding what the evidence actually records. Years in a classroom measure time; a credential records a course of study or a threshold crossed; earnings record a market outcome shaped by the work available and the terms on which it is obtained. None alone supplies an inventory of everything a person has learned. Two courses of equal length can teach different things, and two people with similar skills can have different opportunities to use them. These distinctions matter when an individual calculation is enlarged into a national growth claim. Showing that an educated worker earns more does not by itself establish what would happen if education changed for everyone, with different hiring standards, occupations, tools and wages. The proposed mechanism must survive that enlargement. Evidence of learning, productive use and effects on others can then be joined to the earnings comparison, with each observation doing the job it can support.5
The wealth in a person's capabilities is built through opportunities that the finished credential can hide. Someone taught the first lesson, provided time to practice, kept the equipment usable, or made it possible to persist when learning did not yet pay. Those conditions help turn potential into competence; the word talent can obscure their contribution if it makes every accomplishment sound like an initial endowment. An economy seeking to enlarge what its people can do must therefore attend to both acquisition and use: learning worth the time it takes, routes into productive work, and institutions through which knowledge can travel. The machines remain necessary companions. So do purposes for education that an earnings calculation cannot price. Human capital names one consequence of developing a person, and leaves the person larger than the account. Growth gains a deeper foundation when it recognizes how much productive capacity has to be cultivated before it can be employed.