Training for one career, working in another – new study uncovers the cost

Training for one career, working in another – new study uncovers the cost

The German dual-apprenticeship system is highly regarded around the world as an alternative to university education for young people who would prefer to start working while they learn.

But is there a risk to committing to a long-term occupation when you are only in your teens?

In a recent paper published in EconometricaLink opens in a new window, Dr Dita Eckardt explores what proportion of apprentices stay in the occupation for which they trained, what costs are incurred by those who change their minds, and what could be done to reduce these.

Dr Eckardt explains:

“While Germany’s apprentice system is praised for preparing young people for work with high quality occupational skills training, there is a potential risk – apprentices may find themselves locked onto a career path they chose when they were still very young.

“As we move forward in our working lives, we often update our understanding of the skills we like to use or the fields we want to work in. There is also the risk that developments in technology may make whole occupations obsolete, forcing workers with specialist skills to find new careers.”

Dr Eckardt’s paper offers the first analysis of the scale and costs of this lack of flexibility.

Drawing on a database of German social security records from 1975 to 2010 to create a training/occupation matrix with 13 occupational categories, she finds that at any given point in time, 40 per cent of workers are in occupations that they did not train for.

Establishing the causal wage effect of this type of occupational mismatch is not straightforward. For instance, workers who decide to work outside their training field may do so because they realise they have a particular talent in a different field, and they may earn higher wages in this new field because of their talent, regardless of their original training.

To overcome this challenge and estimate the costs of mismatch, Dr Eckardt combined the occupation data with information on apprenticeship vacancies posted by employment agencies. Because lower demand in one field may push people to work in another for reasons that are unrelated to their talents, fluctuations in vacancies make it possible to isolate the causal effect of occupational mismatch.

The analysis found that the average costs of mismatch were a wage penalty of 14 per cent, or the equivalent of losing two years’ work experience.

The costs become smaller with time, suggesting that workers catch up by learning on the job in their new occupations, but are larger if the new occupation is radically different to the one the worker trained for.

Why do so many people end up working in occupations they did not train for? Dr Eckardt’s analysis suggests that young people may not hold enough information on their true talents, interests, or future labour market developments when making their training choice.

She explains:

“We know from other work that such information problems tend to be more pronounced for students from less advantaged backgrounds. The relatively young age of apprentices compared to university graduates may also be a factor.”

What could policymakers do to tackle the problem? One option would be to improve the information provided to young people before they choose their apprenticeship, but it is difficult to know how much mismatch this would prevent. Another option would be to offer retraining programmes for apprentices who want to switch paths. Dr Eckardt calculates that the up-front costs of retraining would be recovered by a majority of workers and recommends that policymakers should expand retraining schemes.

The paper’s findings carry lessons that extend beyond Germany. Many countries are considering expanding their apprenticeship training systems and Dr Eckardt’s research suggests that policymakers should pay close attention not only to the quality of specific skills provided, but also to the flexibility that is allowed for within those systems.

• Training Specificity and Occupational Mobility: Evidence From German ApprenticeshipsLink opens in a new window (2026) Eckardt, Dita. Econometrica, Vol. 94, No. 3 (May, 2026), 741–766