Paying for job training has always been a challenge. When a Colonial American "master" shoemaker took on an apprentice, it would cost him at the outset. The master had to feed and clothe the apprentice as well as spend significant amounts of time training him. Somehow the master needed to be compensated, apprentices rarely had the funds to pay the full cost of the training upfront. Instead, the master was usually compensated in the form of a commitment to remain a certain number of years. During the early part of the apprenticeship, the apprentice would be more trouble than he was worth. During the later part, he was essentially paying the master back for his trouble.
What happened when an apprentice skipped out in the middle of his apprenticeship? He would be jailed, of course. A deal's a deal. Newspapers during the colonial period were full of runaway apprentice ads alongside notices about runaway slaves.
We don't jail runaway apprentices anymore. (Nor do I say we should!) But there is a price to be paid for refusing to do so. We still have plenty of young Americans without the funds to pay for their own job training. But since employers can't force their trainees to stay till they've have paid off their "debt," they can't be assured they will be compensated for their efforts. Consequently, traditional apprenticeships are much more rare today than they were in an earlier era.
The dominant solution to the "runaway apprentice problem" in America today is "socialized education" (otherwise known as public education). Employees of the state provide various sorts of job training. At the high school level, it is free to the student. At the community college and university level, it is heavily subsidized. Government guaranteed credit is available to cover the rest.
But every solution brings with it its own problems. The downside to public education is that the incentives for the state-salaried instructors are highly imperfect. Schools don't have access to the best information about what the needed skills of the future will be. That information is spread among thousands and thousands of large and small businesses. So schools can end up training lots of travel agents, at a time when travel agent jobs have dried up. Even when they have the information, they may be sluggish to respond to it. Since their pay comes out of tax money, not out of profits, the incentive get things right is blunted.
Germany has a different approach. More than half of young Germans serve at least one apprenticeship. And Germans are less likely than Americans to go to college. But because German employers provide the training, they face the same problem that early American "masters" who took on apprentices had: How can you ensure that the apprentice won't skip out on the apprenticeship?
The dominant solution to the "runaway apprentice" problem there is to have complicated certification requirements for many jobs. If you don't stick it out in your apprenticeship, you won't get your certification and you therefore won't get a job in your chosen field. Union membership can work the same way there. If you don't finish your apprenticeship, you don't get into the union (and hence you don't get a job).
Americans aren't keen on unions having that level of power. Many have fought hard for right-to-work laws and aren't about to back away for those laws. As for licensing and certifications, there is now bipartisan agreement: We already have too many licensing and certification requirements. Most Americans don't want more.
As a result, we probably can't hope to have the large number of apprenticeships the Germans have. When a German-style apprenticeship program in the construction industry was tried in North Carolina a while back, the inevitable occurred: Apprentices skipped out on their employers part way through their apprenticeship and went to work for companies that weren't participating in the program. The latter could afford to pay higher wages, since they weren't paying for on-the-job training.
None of this means that apprenticeships will never work in this country or that they aren't needed. Our current methods of teaching marketable skills generally leave a lot to be desired (and even if they were excellent, they aren't everyone's cup of tea). But it's hard. I wrote about all this in an essay entitled Apprenticeships: Useful Alternative, Tough to Implement. In it, I tried to suggest some partial solutions to the problem.