Alexander Gehret, Andreas Kuhn and Jurg Schweri, Is Training Apprentices Worthwhile? Company Perspectives on Costs, Benefits and Training Quality 2025, Swiss Federal University for Vocational Education and Training SFUVET, Zollikofen, November 2025, English edition
Source ID
SRC-269
Citation
Alexander Gehret, Andreas Kuhn and Jurg Schweri, Is Training Apprentices Worthwhile? Company Perspectives on Costs, Benefits and Training Quality 2025, Swiss Federal University for Vocational Education and Training SFUVET, Zollikofen, November 2025, English edition
Institutional research report, commissioned by the Swiss State Secretariat for Education, Research and Innovation
Hierarchy tier
1
Published
2025-11
Accessed
2026-08-19
Snapshot
Retrieved 19 August 2026, HTTP 200, 1,870,805 bytes, 70 pages. Executive summary, chapter 3.3 on the training rate, chapter 4.1 and chapter 4.11 on the comparison with Germany read at the page. research/ is not committed
Claims
CH7 the Swiss apprenticeship evidence and its transfer limits
Status
READ_IN_PART
Version history
The register keeps changing after the book is fixed in print. Each entry below was added to this row in the order shown.
RETRIEVAL 9 OF THE CHAPTER 7 GATE. The brief retains the Swiss apprenticeship material 'with its transfer limits', which means the underlying evidence is read rather than carried from the earlier book. THE SOURCE MANUSCRIPT'S CLAIMS AT LINES 1651 TO 1665 WERE READ FIRST, AT THEIR LINES, AND ARE CHECKED AGAINST THIS DOCUMENT ONE BY ONE. CONFIRMED. The fifth survey, commissioned by SERI, covering the 2022/23 training year. Responses from almost 6,700 training companies and more than 3,600 non-training companies, which is the figure the ministry's own summary page gives; THE REPORT ITSELF SAYS 'more than 6,600 training companies', so cite one or the other and not both as though they were the same number. VERBATIM ON THE HEADLINE: 'A majority of apprenticeships (around 71 per cent) therefore achieved a net benefit for the training company.' The source's 'over 70 per cent' is right. THE AVERAGES, PER APPRENTICESHIP YEAR, IN SWISS FRANCS: gross costs 27,990; productive output 32,530; NET BENEFIT 4,540. By programme: 9,630 for the two-year Federal VET Certificate, 13,940 for the three-year Diploma, 17,510 for the four-year Diploma. THE GERMANY COMPARISON IS NOW QUANTIFIED, WHICH THE SOURCE ONLY ASSERTS. Table 14 at page 40, German figures from the BIBB survey for the same 2022/23 year converted at the 2022 average rate, per apprenticeship year in CHF: gross costs Germany 26,085 against Switzerland 27,984; wage costs 15,732 against 13,923; personnel costs 6,440 against 11,847; PRODUCTIVE OUTPUT 18,037 AGAINST 32,526; NET BENEFIT MINUS 8,047 IN GERMANY AGAINST PLUS 4,542 IN SWITZERLAND. So the source's 'substantial net costs' in Germany is right and the magnitude is CHF 8,047 of net cost per apprenticeship year against CHF 4,542 of net benefit. THE MECHANISM IS HALF CONFIRMED AND THE OTHER HALF IS NOT THIS DOCUMENT'S. The source says the published explanation is 'two design choices: Swiss apprentices spend more of their time on productive work rather than in simulated training, and Swiss apprentice pay is lower relative to the skilled wage for the same occupation.' THE FIRST IS THE REPORT'S OWN WORDS: productive output is 'about 80 per cent higher in Switzerland than in Germany. One reason for this is that Swiss apprentices spend less time on training exercises and more time on productive work.' THE SECOND IS NOT STATED HERE. What this report says about pay is compositional: apprentice wages are 60 per cent of training costs in Germany against 50 per cent in Switzerland, and in absolute francs Swiss wage costs are the lower of the two. IT DOES NOT SAY SWISS APPRENTICE PAY IS LOWER RELATIVE TO THE SKILLED WAGE FOR THE SAME OCCUPATION, and it attributes the structural difference to 'stricter regulation of the German labour market', citing Muehlemann et al. 2010. SO THAT LIMB NEEDS ITS OWN CITATION BEFORE IT IS PRINTED, and Muehlemann et al. 2010 is the named route to it. A TRANSFER LIMIT THE SOURCE DOES NOT CARRY AND THE CHAPTER SHOULD. THE SWISS SYSTEM RESTS ON A MINORITY OF FIRMS. In the target population there are 53,828 training companies and 273,481 non-training companies, 327,309 in all, giving an aggregated training rate of about 16.5 per cent: 'around one in six companies in Switzerland trained at least one apprentice during that year', among companies with at least two employees. The report adds that some non-training companies 'would not be approved to provide training because they are too specialised'. A system in which five firms in six do not train is a different proposition to transplant than one described only by its net benefit. WHAT PROSE MAY SAY: the headline, the averages, the Germany table with both signs, the productive-work explanation in the report's own terms, and the 16.5 per cent training rate. WHAT PROSE MAY NOT SAY: that this report explains the gap by relative apprentice pay; that the Swiss net benefit is a return on public money, since it is a company-level cost-benefit result and not a fiscal one; or that any of it forecasts a Syrian outcome, the survey being of Swiss firms in a Swiss labour market. THE SOURCE'S OWN LIMITS AT LINE 1663, that Syria lacks the employer associations, the formal private sector and the century of social partnership, ARE UNAFFECTED BY THIS READING AND ARE STRENGTHENED BY THE 16.5 PER CENT. ONE THING TO CHECK BEFORE ANY TREND SENTENCE. The 2016/17 predecessor appears in search results to report 'a majority of COMPANIES (over 60 per cent)' while this survey reports 'a majority of APPRENTICESHIPS (around 71 per cent)'. IF THAT IS RIGHT THE UNIT CHANGED BETWEEN SURVEYS and 60 against 71 is not a trend. IT IS NOT VERIFIED: the 2016/17 report has not been opened and this row records it as a question, not a finding. Route: Gehret et al. 2019, cited throughout this report. STILL UNREAD: chapters 5 to 8, including the multivariate analyses and chapter 7 on in-house training quality