Haute Lumière
Commerce · II.09 · MMXXVI · daylight
ASSESSMENT · Chapter II.09 — Evolution and Economic Selection
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. Write the Price equation and name each of its two terms.
Δz̄ = Cov(w,z)/w̄ + E(w·Δz)/w̄. The first is the selection term — the covariance between the trait and offspring per parent. The second is the transmission term — how faithfully those who persist pass the trait on. One mark for the expression, one for naming both terms. A strong answer adds that there is no third place for a change in the mean to come from.
2. In Nelson and Winter's framework, what is the unit of inheritance in an economy, and why is it not the firm?
The routine — the repeatable, largely tacit pattern by which a thing is done. Not the firm, because firms die while their routines travel: carried by people who move, copied by suppliers, inherited wholesale by spin-outs. A routine can outlive every entity that ever held it.
3. State Wilson and Sober's five criteria for group selection to operate.
(i) The population is structured into groups that persist and that form and disperse in repeated cycles; (ii) groups differ in the trait — between-group variance is strictly positive; (iii) the trait affects group fitness; (iv) group composition is heritable, so offspring groups resemble parent groups; (v) between-group selection exceeds within-group selection. Half marks for four. Criterion (ii) is the one most often left out and it is the one that fails first in practice.
4. What are the three requirements for anything at all to evolve, and what is the diagnostic use of the list?
Variation, selection, inheritance. The diagnostic: when a population is not improving, exactly one of the three is usually missing, and each has a different fix — widen the variation, sharpen the criterion, build the transmission.
Four on application.
5. A group scorecard rewards this year's operating margin. The strategy document commits the firm to a decade of capability building. Which one governs the population of routines, and why?
The scorecard. It is the fitness function — it determines which units carry weight into the next period, and therefore what the covariance term does. In the chapter's worked population the same five firms move at +0.6545 points a generation under a revenue measure and −0.5688 under a margin measure. Full marks require naming the covariance as the mechanism, not simply saying "incentives matter."
6. Your industry body proposes merging the six regional operators into one, citing duplication. Give the evolutionary objection, in its exact form.
Consolidation collapses
V_between. From the thresholdb/c > 1 + V_within/V_between, asV_betweenfalls the required ratio rises without bound, and group-level selection becomes impossible. More plainly: you are removing the only evidence that another way of operating works. Credit any answer that also names the cost as informational rather than competitive.
7. A colleague explains the last decade of your industry by naming a trait the survivors shared. What is the specific defect in this account, and what would fix it?
Fitness has been inferred from survival, so the claim is unfalsifiable — "the fit survived" and "the survivors survived" are the same sentence. The fix is to operationalise fitness independently of the outcome it explains (a defect rate, a cost per unit, a repeat-purchase rate) and to state a predicted rate with an interval before the next period runs.
8. Muir's group-selected hens improved on every measure that mattered. What exactly was changed, and what was not?
The unit of selection — from the individual bird to the cage. Nothing was done to the birds; the same genetic material was scored against a different boundary. Mortality fell from 68.0 percent to 8.8 percent and eggs per hen housed rose from 91 to 237 over six generations. The strongest answers transfer it: the equivalent corporate move is a change to the reporting boundary, not to the people.
Two that require the arithmetic to be done.
9. Six firms sit in two clusters of three, with z = 6, 8, 10 and z = 0, 2, 4. Individual fitness is w = 1.0 + b·Z_k − c·z, with c = 0.030. Compute the critical b above which the trait rises, and say what happens at b = 0.020. Show your working.
V_betweenis the variance of the cluster means 8.00 and 2.00 about 5.00, soV_between= 9.0000.V_withinis the mean of the two within-cluster variances, each(4+0+4)/3, soV_within= 2.6667. The trait rises when(b−c)·V_between > c·V_within, i.e.b/c > 1 + 2.6667/9.0000 = 1.2963, sob = 0.030 × 1.2963 =0.03889. Atb= 0.020 the ratio is 0.6667, below the threshold, and the global selection term is −0.1789 points a generation — the trait is selected out with no change to any firm. Credit a correct threshold even where the final term is only signed rather than computed.*
10. A routine holds a 3 percent annual selection advantage and 2 percent of the market. How long until it holds half the market? Then compute what advantage would be needed to get there in twenty-five years.
t = ln[(0.5/0.5)·(0.98/0.02)] / s = ln(49.00)/s = 3.8918/s. Ats= 0.03,t= 129.7 years. For 25 years,s = 3.8918/25 =0.1557, so roughly a 15.6 percent annual advantage — near the chapter's 0.15 case, which takes 25.9 years. The point of the question is that "the market will sort it out" has a clock on it, and the clock is usually longer than a career.
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.
- What are you actually being selected on? Not the stated objectives — the number that determines whether your work carries weight into next year. Write it in one sentence, then write how long you have known it.
- Name one thing you do well that no measure you are scored against can see. What would have to change in the accounting for it to become visible — and would you want it to be?
- Where have you been part of a group that lost collectively while every member won individually? What would the group term have had to be worth for that to reverse?
- Which of your own routines could survive your departure next month? Write the list. Then write what it would take this quarter to add one more to it.
- Think of an idea you carried out of a previous organisation. What did you keep, what did you lose in the carrying, and what does that tell you about your own transmission fidelity?
- What have you explained after the fact with a story that would have fitted the opposite outcome equally well? Write the version of that claim that could have been wrong.
- Where are you the last person doing it differently? What has that cost you, and what would be lost if you stopped?
- What number would you be willing to predict for the coming year, with an interval, and be scored against? Write it down and date it.
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. Penrose's objection. Edith Penrose argued in 1952 that biological analogies have no place in the theory of the firm, because firms act with foresight and organisms do not. Argue either that the chapter's prospective-rate test answers her objection fully, or that foresight genuinely breaks the analogy in a way no test can repair. Use Penrose (1952) and Nelson and Winter (1982), and one source on strategic intent or managerial cognition that the chapter does not cite.
2. Group selection and the corporation. Wilson and Sober's five criteria were formulated for biological populations. Take a position on whether a corporate group — divisions within a firm, or firms within a supply network — actually satisfies criterion four, group-level heritability, or whether corporate groups reassemble so freely that offspring groups never resemble parent groups. Use Sober and Wilson (1998), and one source on organisational imprinting or founder effects that the chapter does not cite.
3. The measure that makes the population. The chapter claims an economy gets the firms its accounting rewards. Argue the counter-case: that accounting standards are a weak and lagging influence compared with product markets, capital markets or regulation, and that the chapter has mistaken a scorekeeper for a selector. Use the chapter's Price equation material, and one source on the history of a specific accounting standard and its behavioural consequences that the chapter does not cite.
4. Selection versus design. Replicator arithmetic puts a better routine with a 3 percent advantage at 129.7 years to half a market, while deliberate copying can do it in a decade. Argue whether the correct response is to accelerate selection — more variants, harder criteria, shorter generations — or to abandon selection as the mechanism and build transmission institutions instead. Use Dyer and Nobeoka (2000) and Garud and Karnøe (2003), and one source on diffusion of innovations that the chapter does not cite.
5. Is an identity ever evidence? The Price equation holds for every population, including one under no selection at all. Argue either that this makes it a genuine accounting framework of the kind physics calls a conservation law — indispensable precisely because it is always true — or that its universality makes it rhetorically dangerous in economics, where it lends the appearance of mathematics to claims that carry none. Use Price (1970) and Frank (1995), and one source from the philosophy of science on tautology and explanatory power that the chapter does not cite.