Haute Lumière

Commerce · V · MMXXVI · daylight

La Bourse  /  Volume V  /  Nº V.E3

The Translation: One Economy, Two Vocabularies

Volume V — Labour, Value, Flourishing · Extension III of III Two vocabularies, one payroll.

---

THE PLATE

---

THE LETTER

The eleven chapters of this volume were written in two languages at once, and neither of them announced itself.

One is the language of the firm: headcount, turnover, engagement, wage, benefits cost, training expense, absence, overtime, the account balance, the return on investment. Every term in it is a line somebody signs. It is precise, it is auditable, and it has been refined for a century by people whose professional survival depended on getting it right.

The other is the language this house writes in: capacity, depreciation, condition, claim, maintenance reserve, capital formation, recovery, the stock behind the flow. Every term in it names something that is physically there, whether or not anybody books it.

This chapter's whole argument is that these are not rival descriptions. Each row of the table below is one object, and the arithmetic that carries one name into the other is printed beside it — because a translation table without an operation is a glossary, and a glossary proves nothing.

Then the harder half, which for this volume is the richest of the seven. There are concepts in one vocabulary with no counterpart at all in the other, and the language of the firm has no word for most of what Chapters V.04 and V.08 spent their Arithmetic movements measuring. There are fourteen of them below.

And there is a finding in that list which we did not expect and which reorders the rest: every single missing word is on the far side of one of exactly two boundaries. Not fourteen different absences. Two, and the fourteen are where those two fall. Which means the answer is not a new vocabulary. It is two lines moved, and this volume's eleven instruments are already, without ever saying so, almost nothing but those two moves.

— The Editors

---

DISCOVERY

Translations that are already running, with a ledger attached

Begin with the ones that work, because there are more of them than either vocabulary's partisans usually admit, and every one of them was built by somebody with a statutory duty to be correct.

The household satellite account. The Office for National Statistics values unpaid household service work and publishes it beside the national accounts. For 2016: £1,240 billion against a UK GDP of £1,968 billion — 63.0 per cent of recorded output, produced entirely outside it — of which childcare alone was £352 billion, 28.4 per cent of the unpaid total and 17.9 per cent of GDP. That is a national statistician performing exactly the operation this chapter is about: taking an object that exists in one vocabulary and only one, and giving it a number in the other. Nothing was invented. Hours were counted and priced.

IFRIC 2. When a cooperative issues shares to its members, are they equity or a liability? The accounting profession did not resolve this by argument. It wrote an interpretation that turns on one operational fact — whether the entity holds an unconditional right to refuse redemption — and the classification follows from the rules the cooperative actually wrote. Chapter V.06 turned that into a drafting instruction. A translation with a rule is a translation somebody can act on, and that is the difference between IFRIC 2 and every manifesto about the cooperative form.

The WELLBY. HM Treasury's supplementary Green Book guidance values a one-point change in life satisfaction held for a year at £13,000. Whatever you make of the figure, it puts loneliness, noise and job security into the same appraisal arithmetic as a bypass. Before it existed, a scheme whose entire benefit was that people felt better had no column to be written in, and things with no column do not get built.

NICE's published threshold. A stated cost per quality-adjusted life year did something an unpublished rule cannot: it made the rule arguable. Because the number was on paper, Karl Claxton and colleagues could measure what the health service actually displaces at the margin and report £12,936 — against a published upper threshold of £30,000, which exceeds it by 2.32 times. The translation was improved because it had been written down.

The experience modifier. An insured payroll of $40,000,000 at a manual rate of 0.025 produces a manual premium of $1,000,000; a modifier moving from 1.05 to 0.80 is worth $250,000 a year. That is injuries — a condition of a stock — converted into a premium, audited by a third party, every year, already, without anybody having to be persuaded of anything. Chapter V.01 used it as a repayment mechanism precisely because the translation already existed.

Denmark's Employers' Reimbursement Fund. DKK 2,821 per full-time employee, paid by every employer whether or not it trains, reimbursed only to those who do. An externality that economists have described for sixty years, priced, in a single per-head figure. It is not appealed to. It is charged.

And the Swiss training ordinance. An apprentice's time in the firm, the share of it spent on productive work, and productivity relative to a qualified worker: 0.79 × 0.77 × 0.40 = 0.2433 in year one. Three numbers written into a national standard by the professional association for that occupation, which is what makes an apprentice's contribution verifiable across firms and therefore makes Chapter V.05's payback month a number a bank could lend against.

The pattern, across all seven. In every case the translation happened because somebody wrote one number, in one document, that both vocabularies had to accept. Not a reconciliation, not a framework, not a dictionary. A figure with a method attached, published, and then argued about. That is the entire mechanism, and everything below is an attempt to do it thirteen more times.

---

THE ARITHMETIC

One object, two names, and the operation between them

Each row states the firm's number, the stock's number, and the operation that carries one into the other. The worked employer throughout is 5,000 people at a median tenure of 3.90 years, which is the employer Chapters V.09 and V.E2 work.

1 · Headcount ↔ capacity. The firm counts people at an instant. The stock has a quantity and an age.

  headcount                                    5,000 people
  mean tenure                                   3.90 years
  accumulated person-years of capability      19,500
  separation hazard  ln 2 / tenure          0.177730
  people replaced in a year                      889

Now let two hard years take mean tenure to 2.70:

  accumulated person-years                    13,500
  person-years lost                            6,000    =  30.77 %
  change in headcount                           0.00 %

Thirty per cent of the capacity is gone and every report in the building says the number is unchanged. Headcount is the stock's count; capacity is the same stock weighted by how long it has been accumulating. One object.

2 · Turnover ↔ depreciation. The firm records a replacement cost per leaver. The stock is losing value two ways at once, and the two multiply. Chapter V.05's three measured depreciation rates, combined with a 20 per cent separation rate as 1 − (1 − δ)(1 − s):

  delta      half-life      combined      combined half-life
  ----------------------------------------------------------
   1.35 %      51.34 yr       21.08 %            2.93 yr
   3.04 %      22.80 yr       22.43 %            2.73 yr
  10.41 %       6.66 yr       28.33 %            2.08 yr

The three published deltas span a factor of 7.71, and yet the combined rates barely differ, because at 20 per cent separation the departures dominate everything. Turnover is the depreciation schedule, and in most firms it is the larger term. A firm with a turnover line and no depreciation line has half the entry.

3 · Engagement ↔ condition. The firm buys a score. The stock has a state.

  true-score correlation                        0.22
  variance explained                            4.84 %
  explained by everything else                 95.16 %
  the top-versus-bottom quartile gap          2.5422 SD of the predictor
  x r                                          0.559 SD

The firm's sentence is the top quartile is 23 per cent more profitable. The stock's sentence is the condition measure moves the outcome by 0.559 standard deviations. They are the same coefficient. The first is the second with its tails cut off.

4 · Wage ↔ claim. The firm's wage is the price of an hour. The stock's wage is the worker's claim on what the hour produced. Chapter V.01's markdown, w = MRP · e / (1 + e):

  e =  1.5    the wage is 60.0 % of marginal revenue product
  e =  3.0                 75.0 %
  e =  5.0                 83.3 %
  e = 10.0                 90.9 %

At e = 3 the claim exceeds the price by 25.0 points, and Chapter V.08's δ of 5.5 to 17.0 per cent sits on top of it. The long series says the same thing from a distance: US net productivity grew 59.7 per cent from 1979 to 2019 while the typical worker's hourly compensation grew 15.8 per cent — a ratio of 3.78. The price and the claim are one object, and the difference between them is not in anybody's ledger.

5 · Benefits cost ↔ maintenance reserve. The firm books a hundred per cent of the cost. Chapter V.09's lease arithmetic says what it owns:

  social discount rate                          3.0 %
  capture  r / (r + hazard)                    14.44 %
  the hurdle multiple                           6.92 x

The firm pays for the whole asset and holds a seventh of it. That is not a complaint about firms; it is a description of a maintenance reserve on equipment the firm rents, and it explains the shape of the entire corporate health literature without reference to anybody's motives.

6 · Training expense ↔ capital formation. Two countries, one object, two entries.

  Swiss cost per apprentice-year           CHF  28,070
  Swiss output per apprentice-year         CHF  31,240
  net                                      CHF   3,170
  German net cost per year                 EUR   3,596
  over three years                         EUR  10,788
  rent held back, per year                 EUR   4,000  over 5 years
  break-even retention                          53.9 %
  measured German retention                     53.0 %

The German model sits 0.9 points the wrong side of its own break-even and holds anyway because its institutions hold mobility down; the Swiss payback lands at month 27.7 of thirty-six, inside the contract. In both countries the accounts record an expense. The object is a capital formation with a payback month and a depreciation rate, and the vocabulary that can see that is not the one in the statutory accounts.

7 · Absence ↔ the stock reporting its own state. The firm counts days lost. Chapter V.02's shift data counts the same thing before it becomes a day:

  hour 9   relative risk                        1.13
  hour 10                                       1.27
  hour 12                                       2.00

  hours nine to twelve add                     50.0 % more time
  and                                         100.0 % more risk

At fifty-five hours a week and above, coronary heart disease risk is 1.13 times and stroke risk 1.33 times, across 603,838 people, and the WHO and ILO attribute 745,000 deaths in a year to that exposure. Absence is not a cost the stock imposes on the firm. It is the stock telling the firm what the last four hours cost, on a lag.

8 · Overtime premium ↔ the rate on borrowed capacity. This is the row where the two vocabularies agree most exactly and neither notices.

  overtime hours                             48,000 h/yr
  straight-time equivalent                $1,344,000
  the premium on the face                   $672,000   =  50.0 %
  injury cost carried                       $100,800
  replacement cost carried                   $51,300
  the effective rate on the borrowing            61.32 %

Sidney Chapman wrote in 1909 that a working day long enough to draw down the worker's capacity borrows from tomorrow's output at an interest rate nobody books. Here it is booked. It is 61.32 per cent, and half of it is already sitting in the payroll ledger under the heading premium.

9 · Account balance ↔ concentration exposure. Chapter V.03's participant:

  the statement says                         $130,330
  worth, in diversified terms                    35.4 cents
  which is                                    $46,160
  the concentration costs                     $84,170
  the two sum to                             $130,330

One object, two prices, one statement, and only one of the numbers printed on it.

10 · Return on investment ↔ selection effect. Chapter V.09's decomposition: 5.60 : 1 on the vendor's arithmetic, 1.12 : 1 once you subtract the gap that existed before the programme, 0.00 : 1 at the randomised point estimate. The first is 5.00 times the second. Same programme, same people, same spreadsheet. The difference is entirely which population the denominator describes.

11 · Employment stability ↔ an insurance premium. Chapter V.06 priced the cooperative's trade: a conventional firm laying off 10 per cent with 60 per cent income retention against a cooperative laying off 1 per cent and cutting 8 per cent:

  expected income, conventional                96.00 % of normal pay
  expected income, cooperative                 91.68 %
  the premium                                   4.32 points
  break-even relative risk aversion             3.446

The stability costs 4.32 points of expected income, which is the definition of insurance and not a defect. The firm's vocabulary calls this job security and puts it in an engagement survey. It is a priced instrument and the price is computable.

12 · The score ↔ a reporting function. Two populations, same item:

  raw means                A 5.80    B 5.40    A - B  +0.40
  an order-preserving map of the rungs
  transformed              A 5.10    B 5.55    A - B  -0.45

Same data, same ordering of the rungs, opposite league table. The score is the respondent's private translation of a life into a number, and the firm's vocabulary has no word for that translation at all — which is the eighth item on the list below.

---

The false friends, and they are the honest negative

Five pairs look like one object and are two. A translation table that includes them is worse than no table, because it lends the authority of arithmetic to a category error.

The value of a statistical life against human capital.

  value of a statistical life              $13,200,000
  present value of lifetime earnings        $1,630,736
  the first is                                   8.09 x the second

They are not two estimates of one quantity. Chapter V.01 spent its entire Arithmetic movement on this: they are five prices for five decisions, and pairing "what a person is worth" with "compensation" is precisely the mistake that makes the whole subject feel obscene. The first row of any people- translation table you are ever shown will be that pair, and it is wrong.

Output per hour against output per worker. d ln(Y/N) = d ln(Y/H) + d ln(H/N) is an identity, so the two terms cannot be translated into each other; they can only be added to a third. Output per hour peaks at or below 49.0 hours a week and output per worker at 60.17 — a gap of 11.2 hours, which is the whole of the hundred-and-fifteen-year argument about working time, conducted by two sides each holding one term.

The correlation against the quartile headline — 0.22 and 23 per cent, already worked above. The naive ratio against the randomised one — 5.60 : 1 and 0.00 : 1. And the QALY against the DALY: one is gained and the other averted, the weights run in opposite directions, and one minus a utility is not a disability weight, because the two are elicited by different methods on different populations however routinely they are swapped.

---

The missing words, and the cut

Now the list this volume is richest in: concepts with no counterpart at all. Fourteen of them, and the language of the firm has a word for none.

Then the thing we did not expect. Every one of the fourteen is on the far side of one of exactly two boundaries. The firm's vocabulary stops at the legal entity and at the accounting period, and every missing word is a thing that lives outside one of those two lines.

  missing words                                    14
  outside the entity boundary                       8
  outside the period boundary                       6
  ratio                                          1.33

The classification is this chapter's own and is printed as an assertion in the module rather than as a measurement. But once it is made, the test of it is not an argument — it is whether Volume V's instruments fall the same way. They do, and the split is cleaner than the one above:

  instruments                                      11
  entity  ·  period  ·  both               4  ·  6  ·  1

Eleven instruments, built independently, in eleven chapters, by eleven arguments — and every one of them is either a widening of the entity or a lengthening of the period, and one is both. Nobody set out to do that. It is what an instrument has to be, if the thing it is reaching for is on the other side of one of those lines.

Which gives the operating instruction of this whole extension, and it is one sentence. You do not need a second vocabulary. You need to move two lines. Widen the entity until the party who holds the value is inside it — the pool, the consortium, the trust, the household, the trade. Lengthen the period until the asset's own clock fits inside it — the payback month, the capture ratio, the horizon credit, the ten-year series.

And the honest negative that the same arithmetic produces, because it is not good news:

  missing words per entity instrument            2.00
  missing words per period instrument            1.00
  the entity gap is wider by                     2.00 x

The larger hole is the one this volume has the fewest instruments for, and it is the larger hole for a reason that is not an oversight: lengthening a period is a decision one organisation can take on a Tuesday, and widening an entity requires agreement with somebody who does not report to you. Four of the fourteen — 28.57 per cent — have no instrument anywhere in Volume V that closes them at all, and they are named in the Destiny movement rather than hidden, because a translation chapter that lists only the rows that translate is an advertisement.

---

DREAM

What becomes ordinary

In the organisation that has absorbed this, nobody speaks two languages, because there is one set of documents and each number in it carries both readings.

The staff-cost schedule has a second column. Beside headcount sits accumulated person-years, and when tenure falls the second column falls with it while the first does not, and the paper says which. Nobody calls it a soft metric. It is a count of years, and a count is the hardest thing in any ledger.

The depreciation schedule has people in it. Not on the balance sheet — nobody is asking anyone to misstate a financial statement, and IAS 38 is not going to move — but in the memorandum schedule that sits beside it, where the firm's skill families carry a rate each, and the fast-depreciating ones are trained short and often and the slow ones are trained deep and long. The word poaching has gone out of use in the building, because a firm whose payback lands inside the contract has no exposure to a departure, and a firm whose payback lands outside it has joined a pool.

Every proposal that runs longer than a year states the share of its benefit this entity expects to hold, and who holds the rest. About a third of the time the answer opens a conversation with a carrier, a scheme, a trade body or a neighbouring employer — conversations that were simply unavailable while everyone was pretending the firm owned the whole asset.

The overtime line is read as a rate. When somebody proposes another quarter of structural overtime the paper carries 61.32 per cent on it, and the discussion is about whether that is a good price for the money, which is a discussion a treasurer can have and a discussion about tiredness is not.

And the annual report has one number on it that the organisation would rather not print — the differential, computed against a skill-matched benchmark, with its method attached, on the page beside the charitable donations. Nobody finds it awkward any more. A number that is named stops being free, and a subsidy with a line in the accounts acquires a committee that argues about its size, which is the only kind of permanence available to anything.

---

DESIGN

How to run a translation so that it holds

The test, and it is short. A row that does not move a number in a ledger within one accounting period is a synonym, not a translation. The twelve rows in the Arithmetic each name the line they move — the staff-cost schedule, the depreciation memorandum, the capital paper, the payroll ledger, the insurance premium, the benefit statement, the claim run. Rows that name no line are vocabulary, and vocabulary is what this chapter exists to avoid producing.

The four moves, in order, and only four.

One — put the second reading beside the first, never instead of it. The Office for National Statistics publishes the household satellite account beside GDP, not in place of it, and Chapter V.04's rule was never to publish one valuation alone: a replacement-cost figure invites the reply that it is priced at a poverty wage, and an opportunity-cost figure invites the reply that it is priced at a wage nobody earned. Published together with the wedge named, the account is honest and the objection is pre-empted. The same rule governs every row above. Headcount stays. Person-years arrives beside it.

Two — name the boundary before you name the instrument. For any finding that will not translate, ask which of the two lines it is outside. If the value is held by somebody who is not this entity, the instrument is a widening: a pool, a consortium, a trust, a revenue-sharing clause, a household covenant. If the value accrues on a clock longer than the period, the instrument is a lengthening: an accrual, a reserve, a payback month, a capture line, a multi-year series. Two questions, and they select the instrument before anybody designs one, which is why Volume V's eleven fall so cleanly into four, six and one.

Three — write the operation, not the equivalence. IFRIC 2 does not say that a member's share is equity. It says what makes it equity, and the test is a fact about a document. The Danish fund does not argue that training has positive externalities. It charges DKK 2,821 a head. The Swiss ordinance does not assert that apprentices are productive; it writes 0.79 × 0.77 × 0.40 into a standard. In every durable translation the operation is in the instrument and the argument is not in it at all.

Four — publish the method, and expect it to be attacked. The NICE threshold became better because it was published and could therefore be measured against what the health service actually displaces — £12,936 against £30,000, a factor of 2.32. A translation that nobody can check is not a translation; it is a claim in a new coat. Expect the first computation to be uncomfortable and the method to be contested. Both are signs it is measuring something.

And the sequencing rule, which saves the most. Do the first two moves whether or not you ever build an instrument. An organisation that can state, for any people number it holds, both readings and which boundary the gap lies across has acquired a permanent capability for the cost of a column. An organisation that buys an instrument without them has bought a structure whose purpose nobody can reconstruct in three years.

---

DESTINY

How it holds, and the four it does not reach

Three things keep a translation alive, and each is a decision about where a number lives rather than about what anybody believes.

It holds when both readings are in one document. Chapter V.11's finding was a layout finding before it was anything else: a stock measure in a people deck and a flow measure in a finance pack will never be seen diverging, because nobody holds both at once. The same is true of every row above. Two columns, one page, or the translation is a private accomplishment.

It holds when the operation is in an instrument. A liability has to be explained to an auditor before it can be reversed. A per-head charge has to be repealed before it stops being paid. A clause in a standard has to be renegotiated before an apprentice's productivity becomes unverifiable again. Irreversibility is not conviction; it is drafting.

It holds when somebody outside the two vocabularies verifies it. The experience modifier is audited by a third party, which is exactly why Chapter V.01 could use it as a repayment mechanism without building a verification regime. A translation verified by the party that benefits from it will drift, and it will drift in the direction that makes the accounts comfortable.

Now the four this volume does not reach, named plainly, because they are 28.57 per cent of the missing list and no instrument in eleven chapters closes them.

Unpaid care hours. 16.4 billion hours a day, 2.05 billion full-time equivalents, about $11 trillion at 9 per cent of world output — which implies a world economy of $122.2 trillion and an hourly rate of $1.84, so it is a floor and not a valuation. Women perform 76.2 per cent of it: 265 minutes a day against 83, a ratio of 3.19, which over a year is 1,612 hours against 505 — a gap of 1,107 hours, 0.81 of a part-time working year, unpaid, every year, for half the species. Chapter V.04's consortium closes the employer's share of the childcare bill, at $1,437 a head of verified saving. It does not touch the 1,107 hours. Nothing here does.

The withdrawal band. A British worker on Universal Credit faces a marginal rate of 67.6 per cent against a top rate of income tax of 45.0 per cent — the band exceeds the top rate by 22.6 points, and it is paid by somebody earning about fifteen thousand pounds. It sits across the entity boundary between two systems neither of which can see the other, and no employer-level instrument reaches it.

The party bearing a permanent divergence. A plant can run 168 hours a week against a single day shift's 40 — a 4.2 times multiple on capital utilisation, obtainable only by somebody being awake at four in the morning, and the International Agency for Research on Cancer classifies that shift pattern as probably carcinogenic to humans. The commuter's case is sharper still: an hour each way requires about 40 per cent more income to leave a person as well off, which on a £45,000 salary is £18,000 over 460 commuting hours — a shadow price of £39 an hour that nobody pays. Chapter V.11 named the party. It did not compensate them, and neither does this chapter.

Human capital on the balance sheet. 64.0 per cent of global wealth, and IAS 38 forbids recognising an internally generated workforce. A purchased customer list capitalises; the crew that took nine years to train does not. The absence here is legislated, which is different from merely missing, and it is the one item on the list where the correct instrument is not a facility at all.

And one divergence that was already settled, in the workers' favour, and is invisible for exactly the reason this chapter exists. Annual hours per worker in the United States fell from roughly 3,100 in 1870 to roughly 1,760 in 2019 — 43.2 per cent given up. At today's hourly productivity and 1870's hours, output per worker would be 1.76 times what it is. That difference is not lost output. It is flourishing already bought and paid for, at enormous cost, by people who struck for it, and no account in the world records a penny of it. It is the largest translation failure in this chapter and it went the right way.

---

DELIGHT

What it feels like

There is a particular pleasure the first time a row closes.

Somebody says turnover is running at twenty per cent, and somebody else says so the skill half-life is two point nine three years, and the second sentence is not a rebuke or a reframe. It is the same sentence with the clock put back into it, and the room can see that immediately, and the conversation moves to what a two-point-nine-year half-life implies about the training design. Nobody had to be converted. A unit was supplied.

Then the better one, which arrives when the two boundaries become visible. A finding that has been circling for months — this matters and we cannot act on it — resolves in about four seconds into that is outside the entity, so the instrument is a pool, and the question stops being philosophical and becomes a question about who to ring. Two lines, moved, and a dozen arguments stop being arguments.

And the quietest pleasure is in the plate. Two sheets, the same figures under different headings, and the totals agree. They were always going to agree. The only thing that ever kept them apart was that nobody had put them on the same desk, in the same light, with somebody willing to check.

---

OPERATIONALIZE THIS

At the level of finance

The instrument: a two-reading schedule, adopted by resolution, with a boundary clause.

It is not a facility and it raises no money. It is the smallest durable object that makes every other instrument in this volume reachable, and a treasurer will recognise it instantly, because it is a disclosure schedule with a definitions appendix.

The structure.

The balance-sheet treatment, honestly. Nothing here is recognised and nothing should be. The person-year column is a memorandum; the depreciation memorandum sits beside the fixed-asset note and inside neither of them; the capture line is a disclosure in a capital paper. IAS 38 is not being argued with. What changes is the definitions appendix, which is where a schedule survives a change of chief financial officer — because the operation is written there, and an operation somebody wrote down can be repeated by somebody who never met them. The one item worth taking to the auditors in the first month is whichever row produces an accrual, since Chapter V.08's Mission Differential Account is an obligation under IAS 19 or ASC 710 rather than a provision, and getting that treatment documented in year one makes every subsequent year uncontroversial.

The counterparty. Internal for the schedule itself. External at exactly the point the boundary clause fires ENTITY, and the counterparty is then already named by the classification: a training pool for a poaching finding, a multi-employer trust for a capture finding, a consortium for a care finding, a household covenant for a concentration finding. The classification is the introduction.

The number that decides it. One ratio, and it belongs at the head of the schedule:

        rows whose operation moved a ledger line this period
      ---------------------------------------------------------   >=  1.0
        rows published this period

A schedule that publishes twelve rows and moves none of them is a glossary, and a glossary is how this kind of work usually dies — admired, quoted, and attached to nothing. Against it, the honest denominator, which is published in the same table: of the 14 findings in this volume that have no counterpart in the firm's vocabulary, 8 lie outside the entity boundary and 6 outside the period boundary; there are 4 entity instruments and 6 period instruments to reach them, so the entity gap is wider by 2.00 times; and 4 of the fourteen — 28.57 per cent — are not reached at all.

The first ninety days.

---

APPRECIATIVE QUESTIONS

Twelve, for a room

Discovery — what is already working

1. Which number in this organisation is already read correctly by both finance and the people function — and who built the definition that made that possible?
2. Where do we already translate a condition into a price without anybody arguing about it — an insurance premium, a warranty provision, a maintenance reserve — and what can that instrument teach us?
3. Think of a term one part of this business uses that another part has a different word for. When did you last hear somebody put the two side by side, and what happened?

Dream — what becomes possible

1. If every people number in our pack carried both readings on the same line, which recurring argument would simply stop?
2. Imagine our staff-cost schedule carried accumulated person-years beside headcount for five years. What would the fifth year show us that we cannot see now?
3. If we could name, in four seconds, whether a finding sits outside our entity or outside our period, who would we be ringing by the end of the week?

Design — what we build

1. Which of the twelve rows could we compute this quarter from systems we already run, and which one would change a decision fastest?
2. What is the one finding we hold that will not translate at all — and which of the two boundaries is it on?
3. Who should own this schedule, given that a number owned by whoever benefits from the answer drifts toward a narrative?

Destiny — how it holds

1. What would have to be true for both columns to still be in the pack when everyone in this room has moved on?
2. Which of the four unreached items does this organisation actually bear, and who here carries it today?
3. If somebody five years from now used one of these translations for a decision it was never meant for, what did we fail to write in the definitions appendix today?

---

WORKS CITED

Arrazola, M. and de Hevia, J. (2004). "More on the Estimation of the Human Capital Depreciation Rate." Applied Economics Letters, 11(3), 145–148.

Bond, T. N. and Lang, K. (2019). "The Sad Truth about Happiness Scales." Journal of Political Economy, 127(4), 1629–1640.

Chapman, S. J. (1909). "Hours of Labour." The Economic Journal, 19(75), 353–373.

Claxton, K., Martin, S., Soares, M., et al. (2015). "Methods for the Estimation of the National Institute for Health and Care Excellence Cost-Effectiveness Threshold." Health Technology Assessment, 19(14).

Deming, D. J. and Noray, K. (2020). "Earnings Dynamics, Changing Job Skills, and STEM Careers." Quarterly Journal of Economics, 135(4), 1965–2005.

Ferrer-i-Carbonell, A. and Frijters, P. (2004). "How Important is Methodology for the Estimates of the Determinants of Happiness?" Economic Journal, 114(497), 641–659.

Folkard, S. and Tucker, P. (2003). "Shift work, safety and productivity." Occupational Medicine, 53(2), 95–101.

Hanushek, E. A., Schwerdt, G., Woessmann, L. and Zhang, L. (2017). "General Education, Vocational Education, and Labor-Market Outcomes over the Life-Cycle." Journal of Human Resources, 52(1), 48–87.

Harter, J. K., Schmidt, F. L. and Hayes, T. L. (2002). "Business-Unit-Level Relationship Between Employee Satisfaction, Employee Engagement, and Business Outcomes." Journal of Applied Psychology, 87(2), 268–279.

HM Treasury (2021). Green Book Supplementary Guidance: Wellbeing.

Huberman, M. and Minns, C. (2007). "The Times They Are Not Changin': Days and Hours of Work in Old and New Worlds, 1870–2000." Explorations in Economic History, 44(4), 538–567.

IFRS Interpretations Committee. IFRIC 2, Members' Shares in Co-operative Entities and Similar Instruments.

International Accounting Standards Board. IAS 19 Employee Benefits; and IAS 38 Intangible Assets.

International Agency for Research on Cancer (2020). Night Shift Work. IARC Monographs on the Identification of Carcinogenic Hazards to Humans, Volume 124.

International Labour Organization (2018). Care Work and Care Jobs for the Future of Decent Work. ILO, Geneva.

Jones, D., Molitor, D. and Reif, J. (2019). "What Do Workplace Wellness Programs Do? Evidence from the Illinois Workplace Wellness Study." Quarterly Journal of Economics, 134(4), 1747–1791.

Kivimäki, M., Jokela, M., Nyberg, S. T., et al. (2015). "Long working hours and risk of coronary heart disease and stroke." The Lancet, 386(10005), 1739–1746.

Manning, A. (2003). Monopsony in Motion. Princeton University Press.

Mishel, L. and Bivens, J. (2021). Identifying the Policy Levers Generating Wage Suppression and Wage Inequality. Economic Policy Institute.

National Institute for Health and Care Excellence (2022). NICE Health Technology Evaluations: The Manual.

Office for National Statistics (2018). Household Satellite Account, UK: 2015 and 2016. ONS, Newport.

Pega, F., Náfrádi, B., Momen, N. C., et al. (2021). "Global, regional, and national burdens of ischemic heart disease and stroke attributable to exposure to long working hours." Environment International, 154, 106595.

Pencavel, J. (2015). "The Productivity of Working Hours." The Economic Journal, 125(589), 2052–2076.

Salomon, J. A., Haagsma, J. A., Davis, A., et al. (2015). "Disability Weights for the Global Burden of Disease 2013 Study." The Lancet Global Health, 3(11), e712–e723.

Sprangers, M. A. G. and Schwartz, C. E. (1999). "Integrating response shift into health-related quality of life research." Social Science & Medicine, 48(11), 1507–1515.

Strupler, M. and Wolter, S. C. (2012). Die duale Lehre — eine Erfolgsgeschichte, auch für die Betriebe. Rüegger Verlag.

Stutzer, A. and Frey, B. S. (2008). "Stress That Doesn't Pay: The Commuting Paradox." Scandinavian Journal of Economics, 110(2), 339–366.

US Department of Transportation. Departmental Guidance on Valuation of a Statistical Life in Economic Analysis.

World Bank (2021). The Changing Wealth of Nations 2021.

Note on figures. Every figure in this chapter is computed in lib/verify/V_E3.py and prints with its inputs, its units and its source; run python3 lib/verify.py V.E3. Figures carried from Chapters V.01 to V.11 are recomputed here rather than quoted, except the concentration pair at $46,160 and $84,170, which are carried as Chapter V.03 publishes them because that chapter's certainty-equivalent ratio is printed to fewer places than it is computed to; the module prints both. The assignment of each missing word and each instrument to the entity boundary or the period boundary is this chapter's own and is printed as STATED in the module; the counts and the ratio computed on it are arithmetic performed on an assertion, and a reader who classifies differently will get different counts and should. The elevated turnover case, the business-unit size, the commuting hours and the part-year denominator in the care gap are stated assumptions, printed as such.