Returning to the Labor component of the four factors of production in our quest to build a worldbuilding model of the economy, we will now discuss human capital.

Human capital, unlike other characteristics of one’s labor such as your health or general attitudes toward work, is often domain-specific. You may be a highly trained surgeon, but that would do you little good if you have to plant crops. You might be an expert at negotiating trade deals, but that doesn’t help you if you are trying to program your thermostat. In general, the value of your human capital depends on, and interacts with, the available opportunities you have to apply that human capital.

There are many kinds of human capital. In our model, we’ll focus on three:

• Training,
• Experience, and
• Interpersonal skills.

Obviously, this categorization is artificial. Distinguishing between training and experience is not always easy or useful. And the development of interpersonal skills is influenced by training, but also by cultural context—both in terms of what is considered proper etiquette in that culture, and in terms of whether the culture encourages values like teamwork, taking responsibility, giving proper credit, and politeness or other values such as saving face, kiss-up-kick-down, dominance, and rigid separation of roles. So interpersonal skills should strictly speaking interact with the “Culture” factor of labor in our model. Nevertheless, as a scaffold for our thinking, we’re still going to use this three-part division.

Experience and interpersonal skills are fairly self-explanatory. (Some people develop particular skill in working with others, in a way that measurably shows up in company performance, and these skills can be learned; see Crucial Conversations: Tools for Talking When Stakes Are High.) In the rest of this post, we will be discussing training in particular, and focusing on a key question that has wide-reaching implications: who bears the costs of a worker’s training?

(Note: “costs” include actual money, but also instruction time and effort, the frictions involved in assigning real work to a trainee, and the like.)

Following Kathleen Thelen’s book How Institutions Evolve, we can talk about three kinds of vocational skills: general, industry-specific, and company-specific. General skills are widely applicable, such as literacy or basic computer skills. Company-specific skills, on the other hand, are only useful within a particular company—how to use a custom inventory system, for example. Finally, domain-specific skills are useful within a specific industry.

Because general skills make a worker more valuable across industries, a worker who gains general skills is more likely (all else equal) to leave her current employment and find a better offer. As a result, employers will generally not want to pay for their employees to gain general skills (all else equal), because even though they would benefit from having skilled employees, those employees are likely to be poached away and the current employers are less likely to keep the benefits of such skills. The employees, on the other hand, will want to pay for general skills because the risk is low: even if their current job goes away, the skills will be useful to many other employers. Employees thus get the benefits of having general skills, and are willing to pay for them (if they can afford it!).

For company-specific skills, companies have a much easier time paying for workers to develop them; the skills only have value within that company, so training your workers makes them relatively less likely to leave. As long as the employer is confident that a worker will remain, and as long as company-specific skills would actually be useful, the employer is likely to pay for such training. The employee himself is relatively less likely to pay for company-specific skills, for that reason.

But when we consider domain-specific training, we have a problem. If the worker bears the cost of his own training, he also runs a relatively high risk that no one will hire him for that job even after he is trained (since it only applies to one industry). If so, the cost of the training will be wasted, since he would not be able to apply the specialized training in a different domain. As a result, the worker will be less willing to bear the cost of his own training unless he had some sort of assurance that the investment would pay off.

Conversely, if the employer bears the cost of training a new employee with domain-specific skills, she runs the risk that the employee will receive all the expensive training and then happily jump ship to a different employer, or strike out on his own, or simply underperform at his new job. The employer will be unwilling to spend lots of resources training employees unless she had some sort of assurance that they would remain with her, and perform up to par.

This is probably why medieval Europe featured long apprenticeships and state-sanctioned professional guilds—apprentices could devote years of their life (but did not have to pay money) to learning a trade secure in the knowledge that their master would employ them (albeit under bad conditions), and the master could invest the considerable effort needed to train an apprentice secure in the knowledge that the apprentice could not run off early and ply his trade elsewhere, because the apprentice could be imprisoned or even executed. The apprentice was locked into his contract for several years, long enough for his master to reap the benefits of his growing skill.

On the other hand, there are significant drawbacks to the apprenticeship system. First of all, the master is taking a big risk—what if you turn out to be really bad even with training, or dishonest, or just unpleasant to be around? Second, the apprentice has to sacrifice many years of his life toiling for someone else—and what if the master is cruel, or incompetent, or just bad at business or teaching? Why not take opportunities to abandon your master and improve your life?

Most of all, an apprenticeship system requires overpowering coercion to work—either from a powerful state that enforces contracts between master and apprentice, if you’re lucky enough to live under one, or else a social milieu that tolerates private violence by masters and guild enforcers against the hapless apprentices. Or perhaps both.

In modern times, we typically use other means, which have varying levels of success and different outcomes. Here, we’ll talk about two models, and call these a “liberal” labor system and an “organized” labor system.

In a stylized “liberal” labor system where workers can move between companies and industries without restrictions, companies have less assurance that they will be able to keep workers around after they have been trained; as a result, companies tend to invest relatively little in workforce training (except for company-specific skills), and workers themselves are encouraged to finance their own training.

Workers, for their part, will therefore tend to invest in general skills that do not depend on a particular employer or industry, as they have a higher likelihood of benefiting from such investment wherever they end up. They will also invest (where possible) in especially valuable skills that are industry-specific (such as computer programming), because the expected return from such investment is still positive even with the uncertainty of the payoff.

But less valuable industry-specific skills (such as trades) will tend to be neglected. Moreover, the skill development of the workforce as a whole will largely depend on the workers’ ability to invest in their own skills. If they lack the funds, the time, or the access to credit, workers will not be able to get all the skills they want. (This is a particular problem at the beginning of your career, when you have no money!) As a result, a “liberal” system will tend to produce a workforce with reasonably levels of general skills and highly valuable industry-specific skills, and a large gap of skills in the middle.

One way to address this gap is for the state to provide free or subsidized education to younger people, especially to fund the development of general skills. Unsurprisingly, in the United States over half of the workforce has college degrees, while only perhaps 35% of German workers do (and many of these are professional degrees, rather than what Americans would recognize as a liberal-arts degree).

Another way is for companies to offer strong incentives for employee loyalty, partly mitigating the poaching problem. Examples include the Japanese system of worker seniority, or the common practice among American startups to offer restricted stock compensation that vests over several years.

By contrast, in a regimented system of long-term employment with few opportunities to switch jobs (what Thelen calls an “organized” system, as one finds in Germany), companies will be assured that they can capture the benefits of training investments; each company will therefore train its employees to the level that the company needs (or thinks it needs). However, workers themselves will tend to underinvest in their own skills; because of the limits on job choice, they will not reap all the benefits of such investment.

As a result, an organized system tends to produce a workforce that has good basic and “middle” domain-specific skills, but lacks skill on the high end. (In Germany, for instance, nearly half of workers have attended vocational schools, often funded by their future employers. Germany also features industry groups that collectively manage worker training, and agreements between companies to manage worker poaching.)

****

The upshot is that a skilled workforce doesn’t spring from the ground fully formed. Someone has to bear the costs of training, and that someone has to be confident that she will reap the benefits of that investment. There are several ways to resolve the resulting problems, but each of them will result in a characteristic pattern of skill development—and such patterns can add texture to your invented societies.

*****

(This post is part of Politics for Worldbuilders, an occasional series. Many of the previous posts in this series eventually became grist for my handbook for authors and game designers, Beyond Kings and Princesses: Governments for Worldbuilders. The topic of this post belongs in the planned second book in this series, working title Wealth [Commerce?] for Worldbuilders, along with some overlap with the planned third book, working title Tyranny for Worldbuilders. No idea when they will be finished, but it should be fun!)