La Bourse / Volume V / Nº V.05 / Workbook — the executive

A woman reading on a daybed, the shadows of palm leaves falling across the wall behind her in late sun. — Plate V.05 · Workbook — the executiveThe Twenty-Eighth Month.There is a month in every apprenticeship when the firm stops paying for the learning and starts being paid by it. In Switzerland it is month twenty-eight of thirty-six, and almost nobody — including the apprentice — knows the date.

WORKBOOK — THE CORPORATE EXECUTIVE

Chapter V.05 · Skill, Mastery, and Time

For the person who owns a P&L and has been told, in a meeting, that the problem with training is that people leave. This workbook gives you the arithmetic that answers that sentence, the instrument that follows from it, and the board paper that carries both. Nothing here requires a change in your firm's values. It requires a schedule your accounts do not currently produce.

THE PREMISE, STATED COMMERCIALLY

Your firm holds an unscheduled asset class of material size. It is expensed on acquisition, tracked nowhere, depreciated at a rate nobody has measured, and described as "lost" when it resigns. If you treated plant this way your auditors would have something to say about it.

Three facts, all measured, all in the chapter's Works Cited:

Training can be cash-positive during the training period. The Swiss cost-benefit survey for 2016/17, covering more than 5,700 training firms, puts gross costs at CHF 28,070 per apprentice-year against CHF 31,240 of productive output. A net benefit of CHF 3,170 a year, before anybody is hired.
Or it can be a five-year recovery, and barely solvent. German firms carry €3,596 a year of net cost per apprentice. At 53 per cent retention and a €4,000 annual productivity rent, break-even retention is 53.9 per cent — the model clears by less than a point.
The difference between those two firms is not effort. It is where the payback sits in time.

Your decision is therefore not should we train. It is where does our payback land, and does it land inside the tenure we can count on.

PART ONE — DISCOVERY

Days 1–30: find the number your accounts do not produce

Exercise 1.1 — The training cost reconstruction (one week, one analyst)

Your general ledger does not have a training cost line that means anything. Build one for a single role — the one you hire most.

  per trainee, per year
  ------------------------------------------------------------
  A  trainee compensation, fully loaded            ________
  B  supervisor time x supervisor cost per hour    ________
  C  formal instruction, courses, certification    ________
  D  materials, licences, equipment, rework        ________
  E  recruitment and administration                ________
     GROSS COST = A + B + C + D + E                ________

Line B is the one everyone omits and it is frequently the largest after A. Count it at the supervisor's fully loaded rate, not their salary.

Exercise 1.2 — The productive contribution, honestly (one week)

The Swiss surveys decompose an apprentice's productive value into three measurable factors, and you can use the same three:

  productive value  =  time actually in the business
                       x  share of that time on productive work
                       x  productivity relative to a qualified person
                       x  what a qualified person costs

Swiss three-year anchors, published: year one 0.79 × 0.77 × 0.40; year three 0.81 × 0.84 × 0.80. Relative productive values of 0.26, 0.41 and 0.54.

Get your three factors from the supervisors, not from HR, and get them role by role. Supervisors know the third number to within about ten points and have never been asked for it.

Exercise 1.3 — The tenure number, measured (two days)

Pull your last three cohorts in this role. What proportion were still employed twelve months after completing the programme? Twenty-four?

Use your own data, not an industry figure. German retention is 53 per cent and Swiss is about 35 per cent; the gap between two adjacent countries is eighteen points, so a sector average tells you nothing about your building. Call the measured result T.

PART TWO — THE ARITHMETIC

Days 31–45: compute P, and compare it to T

Exercise 2.1 — The payback curve (half a day)

Lay the two series side by side, year by year, and cumulate.

  year    output      cost        net     cumulative
  --------------------------------------------------
    1     20,138    28,070     -7,932        -7,932
    2     31,756    28,070     +3,686        -4,245
    3     41,825    28,070    +13,755        +9,510

That is the Swiss three-year apprenticeship, using the survey's own aggregate output distributed across its own published productive-value shares. The cumulative line crosses zero at month 27.7 of thirty-six. Under a rising cost profile it crosses at month 26.2. Either way, before the contract ends.

Do the same for your role. P is where your cumulative line crosses.

Exercise 2.2 — The decision (one hour)

      Train on your own book when   P < T
      Pool, or shorten P, when      P > T

That is the whole decision and it fits on one line of a board paper.

Exercise 2.3 — The sensitivity that decides the design (two hours)

Re-run P with the depreciation rate of the skill you are actually building.

  delta        half-life    E(5 yrs) on a 4,000/yr rent at 53% retention
  ---------------------------------------------------------------------
   1.35 %/yr    51.3 yr           10,250 EUR     0.95x cover
   3.04 %/yr    22.8 yr            9,833 EUR     0.91x cover
  10.41 %/yr     6.7 yr            8,264 EUR     0.77x cover

Against a €10,788 three-year investment, the slow-depreciating skill leaves you five per cent short at five years and the fast one leaves you twenty-three per cent short. Over an unbounded horizon the same rents are worth €157,037 and €20,364 — a ratio of 7.7.

The commercial conclusion, stated for a board: rapid obsolescence is an argument against long paybacks, not against training. If your δ is near ten per cent, stop modelling any recovery beyond about seven years and redesign the programme so the contribution arrives earlier.

PART THREE — DESIGN

Days 46–60: the instrument

Four instruments, and your P-versus-T result chooses among them.

One — shorten P. Free, fastest, and almost always available.

The Swiss payback lands at month 27.7 because trainees are in the business four days a week, mostly on productive work, reaching 80 per cent of a qualified person's productivity by year three. Every one of those is a design parameter.

Three levers, in order of effect:

Real work sooner, on a narrower slice. Reaching 80 per cent on one defined task beats 30 per cent across ten, and it beats it in cash.
Instruction inside production rather than beside it. The German surveys show net costs fell by around a third between 2000 and 2007 chiefly because apprentices were used more in real production — and skills improved over the same period. These are complements, not rivals.
Make the contribution visible. Uncounted output does not move P and does not survive the next cost review.

Two — a levy, if you are setting policy rather than paying it.

The UK Apprenticeship Levy: 0.5 per cent of pay bill above £3 million, a £15,000 allowance, a ten per cent top-up, twenty-four-month expiry. A pay bill of eight million pounds pays £25,000 and holds £27,500.

Cost it before you praise it. At 229 employees and the upper measured effect of 0.5 percentage points of training intensity, that £25,000 buys 1.14 additional starts — £21,875 per marginal start, and £109,375 at the low end. National starts fell 24.1 per cent in the year it landed, while the matched evaluation found training rose at levy-payers relative to comparable non-payers. Both are true. A levy is a price signal, and the marginal unit is expensive because most of the money funds training that was already happening.

Three — the bond, and its ceiling.

Do not build your retention case on a repayment clause. German doctrine ties the lock-in to the length of training, two years being the practical ceiling, tapered monthly; English courts after Cavendish Square v Makdessi ask whether the sum is proportionate to a legitimate interest; US regulators treat aggressive versions as employer-driven debt. And §12(1) of the German Berufsbildungsgesetz voids any clause binding an apprentice after training, which removes the instrument in exactly the case where the money is largest.

A bond big enough to matter is unenforceable; a bond the law will keep is too small to matter. Put the drafting time into shortening P instead.

Four — pool, when P > T and cannot be brought inside it.

Denmark has run the reference design since 1977: every employer pays a flat per-head contribution — DKK 2,821 per full-time employee at the 2026 rate, so DKK 564,200 a year for a 200-person firm, about €75,600 — whether it trains or not, and only trainers draw reimbursement. The private equivalent is a group training organisation: a consortium that employs apprentices and rotates them across member firms.

The structure, in the form a treasurer will recognise.

Legal form. The consortium is employer of record and holds the training contract. Members host and pay a hosting fee per apprentice-month, plus an annual subscription per employee whether or not they host.
Balance-sheet treatment. At the member: hosting fees are operating expense in the period, and the subscription is a membership cost. No intangible, no argument with the auditor. At the consortium: the apprentice contract is a fixed-term obligation matched by a receivable from hosting members, with any pre-funded balance carried as a deferred cost recovered over the remaining contract term. Capitalise the contract, never the person.
Subscription pricing. Aggregate shortfall — the cost of the months between T and P across the cohort — divided across total member headcount.
Security. The hosting commitments, and nothing else. If members stop hosting, the consortium stops recruiting; write that in.
Term. Two cohorts, with a hard review after the first completes.

PART FOUR — DESTINY AND DELIGHT

Days 61–90: make it hold, and get it signed

Exercise 4.1 — Put P in the reporting pack (one week)

Anything reviewed monthly persists; anything reviewed by exception evaporates. Two lines, on the standing pack:

  payback month, current cohort (P)        ____ months
  retained tenure, last three cohorts (T)  ____ months

Nothing else. Two numbers on the pack will outlive every presentation you give.

Exercise 4.2 — The failure modes, named in advance (half a day)

Early dissolution. German premature dissolutions rose from 25.0 per cent in 2013 to 29.5 per cent in 2022; Swiss terminations run at 20 to 25 per cent. A departure in month four destroys the entire investment; one in month thirty destroys nothing. Your first-year attrition is worth more management attention than your post-qualification retention, and almost every firm has this backwards.
The life-cycle reversal. The vocational employment advantage is 6.9 points at entry, zero by age forty-nine, and minus eleven and a half points by sixty-five. A firm that trains twenty-year-olds superbly and offers a forty-eight-year-old nothing has moved the problem somewhere nobody measures.
The quiet stop. When mobility rises, firms do not announce that they have stopped training. They take one fewer trainee a year and the sector notices seven years later. The pack catches this; nothing else does.

Exercise 4.3 — Tell the trainees their payback month (one hour)

This costs nothing and changes the first year. It is the same courtesy as telling someone what their equipment cost, and it converts we are investing in you, said with slight embarrassment, into a schedule two adults can both read.

THE BOARD PAPER — one page

Recommendation. [Train on our own book / Form a training consortium with named members], on the basis that P = ____ months and T = ____ months.

The arithmetic. Gross cost per trainee-year ____; productive contribution by year ____ / ____ / ____; cumulative crossing at month ____. Depreciation rate of the skill family ____ per cent a year, half-life ____ years, giving ____ times cover over five years.

The comparable. Swiss training firms run a net benefit of CHF 3,170 per apprentice-year with a payback at month 27.7 of thirty-six, at 35 per cent retention. German firms run €3,596 of net cost a year at 53 per cent retention, with break-even retention of 53.9 per cent. Our position sits [here].

What we are not asking for. No capitalisation of people. No repayment bond. No change to compensation policy.

The decision number. P versus T. Everything else in this paper is how those two numbers were obtained.

Review. After the first cohort completes, against the signed cost and contribution assumptions, by internal audit.

THE FIRST NINETY DAYS ON ONE PAGE

Day	Action	Artifact
1–15	Reconstruct gross cost per trainee-year for one role, line B included	The cost build
16–25	Get the three productivity factors from supervisors, role by role	The contribution build
26–30	Pull three cohorts and measure twelve- and twenty-four-month retention	T, measured
31–40	Build the cumulative curve; find the crossing	P, computed
41–45	Run the sensitivity at your skill family's δ	The decay table
46–55	If P < T, write the memo and stop. If P > T, name three consortium members	Decision memo
56–70	Heads of terms; price the subscription off the aggregate shortfall	Subscription schedule
71–80	Put P and T on the standing reporting pack	Two lines on the pack
81–90	Tell the current cohort their payback month	The conversation

The two artifacts in bold are the ones that survive you. Everything else is how they were produced.

WHAT THIS IS NOT ASKING YOU TO DO

It is worth being explicit, because training proposals usually arrive with freight attached and this one does not.

It does not ask you to capitalise people. You capitalise a contract; the distinction survives audit precisely because it is the honest one.

It does not ask you to change compensation policy. The wage compression that makes firm-financed training viable already exists in your firm, whether or not anybody has named it, and nothing here proposes widening or narrowing it.

It does not ask you to accept a longer payback than you would accept anywhere else in the business. It asks the opposite: apply the same payback discipline you apply to plant, and then notice that under the discipline a large amount of training passes.

APPRECIATIVE QUESTIONS FOR YOUR LEADERSHIP TEAM

Where in this business does someone become genuinely productive far earlier than the job description assumes, and what is different about how that role is taught?
Which of our training designs could have its payback moved six months earlier without teaching less — and what exactly would change in week one?
If we were known as the best place in this sector to be trained and then leave, what would that do to who applies to us?
What would it take for the person we train at forty-eight to get the same quality of formation as the person we train at nineteen?