La Bourse / Volume V / Nº V.11 / Quiz, reflection, essays

A woman in a cream shirt dress standing in a sunlit room, looking toward the windows. — Plate V.11 · Quiz, reflection, essaysTwo Dials, One Hand.Two true instruments can disagree, and the disagreement is data. What you do next depends entirely on knowing which of them measures a flow and which measures the stock the flow is drawn from.

ASSESSMENT · Chapter V.11 — When Flourishing and Output Diverge

Three instruments: a ten-point quiz, eight reflection questions, five essay prompts. The quiz checks comprehension rather than recall. The reflections are private and first-person. The essays are arguable from more than one side.

THE QUIZ — ten points

Four on recall.

1. State the three findings Easterlin put side by side in 1974, and say which two of them are in tension.

Within a country at one moment, richer people report higher happiness; across countries the relationship looked weaker than that gradient predicted; over time within a country, average happiness had not risen with income. The first and the third are the tension: if income buys happiness for a person, why does half a century of it buy so little for a nation? One mark for all three, one for correctly naming the pair in tension. No mark for describing Easterlin as claiming income does not matter — he has never claimed it.

2. State the three properties Stevenson and Wolfers report for the well-being–income relationship.

It is log-linear; it shows no satiation point, including at the top of the distribution in their 2013 follow-up; and the within-country and between-country slopes are similar.

3. Write the chapter's operating rule on stocks and flows.

Output is a flow. Flourishing measures the condition of the stock the flow is drawn from. A flow can always be raised for a period by drawing the stock down, and no flow measure can tell you whether that is what happened. Credit an answer that connects it to S = D / (R · r) from Volume I.

4. Name the three conditions under which a flourishing measure is the artefact rather than the signal.

It is self-reported and not anonymous; it has no anchor and has been running long enough for hedonic adaptation to act; it is in somebody's bonus and states no denominator. A strong answer adds the honest limit: Lucas et al. (2004) found life satisfaction does not return to baseline after unemployment, so adaptation is never a reason to discount a fall.

Four on application.

5. A business unit beats plan by 6 percent. In the same quarter voluntary turnover rises 3 percentage points and overtime rises 18 percent. Diagnose it using the chapter's rule, and say what you would measure next.

Output up, flourishing down — so output is the suspect. The pattern (overtime, turnover) is the signature of a flow raised by drawing the stock down: this is a portion of next year's capacity sold forward and reported as income. Measure next: hours worked against hours contracted, the maintenance and training backlogs, unfilled vacancies, and internal fill rate on senior roles. The stronger answer says the 6 percent is not false — it happened — but that it is recorded in the wrong account, and that the artefact typically expires within four to six quarters.

6. A colleague says: "Well-being rose during the recession, so growth clearly isn't what matters." What is missing?

Two things. First, composition: on Ruhm's elasticity, total mortality falls with unemployment — fewer road deaths, fewer cardiac deaths — while the US suicide rate rose from 11.3 to 12.0 per 100,000 between 2007 and 2009. One aggregate hides both halves. Second, the control case: Russia in the 1990s saw output collapse alongside falling life expectancy and life satisfaction. Falling output is occasionally, partially and accidentally protective, and that is all it is. Full marks require both halves.

7. Your employee-engagement score has risen for three consecutive quarters. The survey is administered by the HR team, results are shared with line managers, and the response rate has fallen from 71 percent to 48 percent. What do you conclude?

That the score should be read as falling, not rising. The measure fails the first artefact condition — self-reported and not anonymous, so it partly measures the relationship with the manager — and a rising score on a sharply falling response rate is the classic signature of the people with the worst answers leaving the sample. Remedies: third-party administration, publish the response rate beside the score every time, and pair it with a behavioural measure that cannot adapt or self-select.

8. Edmans (2011) found a 3.5 percent annual alpha on a portfolio of the best companies to work for. Why is that both the strongest argument in the volume and an argument with a shelf life?

Because an alpha is evidence of mispricing. The employee-satisfaction signal paid precisely because the conventional measures could not see it — the return was the fee for seeing it first. That is a strong argument for building the measure now, and simultaneously an argument that the excess return shrinks as the measure becomes standard. Credit any answer that draws the practical conclusion: the case for measuring is an information case, not a permanent arbitrage.

Two that require the arithmetic to be done.

9. Using the chapter's slope of 0.92 ladder points per natural log unit, how many years of 2 percent real growth buy one full ladder point out of ten? Then compute what the same slope predicts over a 20-year window, and explain in one sentence why both answers are true at once.

Income multiple for one point: exp(1 / 0.92) = 2.9×. Years at 2 percent: ln(2.9) / ln(1.02) ≈ 55. Over 20 years income rises 1.02²⁰ = 1.49×, so the predicted gain is 0.92 × ln(1.49) = 0.37 ladder points — 3.7 percent of the scale. Both are true because the relationship is linear in the log: it is tight, unbounded and slow, so a strong correlation measured over a short window looks like nothing. Credit any method reaching 50–60 years and 0.3–0.4 points. The point of the question is that the Easterlin paradox falls out of the Stevenson–Wolfers slope, and the student should say so.

10. A covenant fires when two KPIs move in opposite directions for two consecutive periods. Compute the false-trigger rate on two independent series doing nothing, first on sign alone and then with a gate requiring both moves to clear one standard deviation. State the improvement factor.

Sign alone: P = 0.5 per period, 0.5² = 0.25 — 25 percent, a quarter of all periods. Gated: one-tailed normal 1 − Φ(1) = 0.1587, so opposite-and-both- material is 2 × 0.1587² = 0.0503, and twice running is 0.0503² = 0.00253 — 0.25 percent, roughly one period in four hundred. The improvement is 0.25 / 0.00253 ≈ 99×. The stronger answer states the conclusion in the right register: the gate is one sentence of drafting and it is the difference between a covenant and a nuisance — and that one standard deviation is a policy choice, while a trigger on sign alone is not defensible at all.

REFLECTION — eight questions, for one person and a pen

These are not for a room. Write the answers by hand if you can; the slowness is the point.

Think of a period when your own output was high and you would not want to repeat it. What stock was the output drawn from, and how long did it take to come back?

Which number about your own life do you trust more than you can justify — and what would you look at to find out whether it is the flow or the stock?

Where have you told yourself that a fall was just adaptation, or just a phase? Lucas and colleagues found that some falls do not recover. Which of yours has not?

Name something you do that creates real value and appears in no measure of your output. What has it cost you that it is uncounted?

Recall a disagreement you resolved by working out that both sides were right over different horizons. What made you able to see that, and where might the same move apply now?

Who carries a permanent cost so that something you rely on can work — a night shift, a commute, an hour of care? Write their name. Not a category. A name.

What would you have to see for two years running before you would change your mind about something you currently believe about work and well-being? Write the threshold now, before the evidence arrives.

If you took 43 percent of your working hours back, as the last century and a half already did on your behalf, what would you do with them — and what does your answer tell you about what you are currently counting?

ESSAY PROMPTS — five

Each is arguable from more than one side. Each requires at least one source the chapter cites and at least one it does not.

1. Is the paradox a paradox? The chapter argues that Easterlin and Stevenson–Wolfers are computing the same coefficient and disagreeing about whether fifty-five years is a long time. Argue either that this dissolves the dispute — that it is an artefact of the observation window relative to a logarithmic relation — or that a genuine substantive disagreement survives the arithmetic, about relative income, social comparison, or aspiration adjustment. Use Easterlin et al. (2010) and Stevenson and Wolfers (2008), and at least one source on relative income or reference-group effects that the chapter does not cite.

2. What should replace or supplement GDP, if anything? The chapter shows that 43 percent of US working hours were given up between 1870 and 2019 and that GDP records none of it, and that unpaid care is worth roughly 9 percent of global GDP and is excluded by construction. Argue either for a composite index, for a dashboard of separate measures kept separate, or for leaving GDP alone and fixing what sits beside it. Use the ILO (2018), and at least one substantive critique of composite well-being indices that the chapter does not cite — Neumayer on the ISEW and GPI is a starting point, not an ending one.

3. Cuba, and the ethics of the natural experiment. Output fell about 35 percent, adults lost 5.5 kg on average, diabetes and coronary mortality fell sharply, and roughly 50,000 people developed epidemic optic neuropathy. Write the case for citing this as evidence about diet and activity, then the case against citing it at all. Conclude with which you find more persuasive and why, and state explicitly what a reader is entitled to conclude from it. Use Franco et al. (2007), and at least one source on the Special Period's public-health record that the chapter does not cite.

4. Who should pay for a permanent divergence? Shift work delivers a 4.2× multiple on capital utilisation and is classified by the IARC as probably carcinogenic when it disrupts circadian rhythm. The premium paid for it is conventional rather than derived. Argue for one of: a compensating differential set by bargaining, a regulated floor on the premium, a reduction in the hours themselves, or a structural shift of the cost onto the capital that benefits. Name who bears the cost under your proposal and who bears it today. Use the chapter's Stutzer and Frey material on uncompensated differentials, and at least one source on shift-work premia or compensating wage differentials that the chapter does not cite.

5. Does measuring flourishing destroy it? The chapter argues that Edmans's alpha exists because the market was mispricing employee satisfaction — that it paid because it was unmeasured. Argue the counter-case: that formalising a flourishing measure will produce the same gaming, narrowing and displacement that every performance measure produces, and that the chapter's three artefact conditions are a patch rather than a solution. Then argue the rebuttal. Use Edmans (2011), and at least one source on measurement effects, gaming, or Goodhart's law that the chapter does not cite.