Haute Lumière
Commerce · II.11 · MMXXVI · daylight
ASSESSMENT · Chapter II.11 — The Case Against This Book
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 — and in this chapter more than any other, an essay that ends where it began has not done the work.
THE QUIZ — ten points
Four on recall.
1. State Popper's test and Lakatos's improvement on it, and say why the second is the more useful instrument for judging a school of economics.
Popper: a theory is scientific in proportion to what it forbids. Lakatos: a research programme is progressing if it predicts novel facts that turn out, and degenerating if it adds assumptions only to absorb results it did not predict. Lakatos is more useful here because no single observation will retire a paradigm, and the live question is whether the programme has told anyone something new and been right. One mark for each test, one for the reason.
2. Name the four columns of the claim register and the three rules that keep it honest.
Columns: claim, class (F or U), kill condition, status. Rules: a U-claim may stay but may not be cited as evidence or appear in a term sheet; a kill condition may be tightened but never loosened, with versions kept visible; the reader of the register is not its author.
3. State Hansmann's explanation for the rarity of worker ownership, and give the figure the chapter reports for the United States.
The binding cost is collective decision-making, which rises steeply with the heterogeneity of owners' interests — not capital access and not hostility. Roughly 1,000 firms and 10,000 workers against 160 million employed: 0.00625 percent, or 6.25 workers in every hundred thousand.
4. Distinguish relative from absolute decoupling, and territorial from consumption-based emissions.
Relative: emissions per unit of output fall. Absolute: emissions fall while output rises. Territorial: emitted inside the border. Consumption-based: emitted anywhere to serve domestic demand, imports included. Full marks require noting that the pairs are independent — a country can show absolute territorial decoupling and much weaker consumption-based decoupling at once.
Four on application.
5. A colleague defends a proposal by saying "the economy is a living system, so the regenerative option is the right one." Diagnose the move, and say what you would ask for instead.
The sentence is a U-claim: it forbids no observation, so it cannot function as evidence, and the rule is that it may not be used to answer an objection. Ask instead for the F-claim underneath it — the crossover year at stated rates, or the measured regeneration rate of the stock in question — and for the finding that would retire it. Credit strongly any answer noting that the U-claim may be true and may stay in the document; it simply cannot carry weight.
6. Someone cites "we are using 1.7 Earths" as evidence that resource limits are binding across the economy. What is wrong with the inference, and what survives it?
About 60 percent of the footprint is the carbon component; strip it and the rest of the human enterprise is 0.684 Earths, inside one planet. The index is therefore a carbon restatement in hectares, and it is not independent evidence about crops, timber, fish or land. What survives is the carbon finding itself, which has three independent physical checks — Mauna Loa, ocean pH, ocean heat content — and needs no composite index at all.
7. Your organisation devolves governance of a shared resource to "the community." Name the public-choice objection and two of Ostrom's design principles that address it directly.
Olson: concentrated interests organise and diffuse ones do not, so a smaller arena is cheaper to capture; Bardhan and Mookherjee found decentralisation helps where local accountability is strong and harms where it is not. Any two of: clearly defined boundaries; collective-choice arrangements including those affected by the rules; accessible low-cost conflict resolution; minimal recognition of the right to organise; nested enterprises. The stronger answer observes that Ostrom's principles are, read together, a capture-resistance specification.
8. A validation function has reviewed forty claims over three years and overturned none. Your head of programme presents this as evidence the paradigm is sound. Respond.
It is evidence of nothing, because a detection probability near zero produces exactly this record whether or not the claims are true. Under the chapter's inequality, a function below a 33.3 percent detection rate does not pay for itself. Ask what the validator's reporting line, budget and standing are — the three components of effective challenge — before reading the result at all. A register with no retirements after five years is an unread page, not a strong theory.
Two that require the arithmetic to be done.
9. A national account reports a footprint of 2.10 Earths with a carbon component of 64 percent. Decompose it. What can and cannot be concluded about non-carbon resource use? Show your working.
Carbon component: 2.10 × 0.64 = 1.344 Earths. Everything else: 2.10 × 0.36 = 0.756 Earths. Overshoot above one planet: 1.10 Earths. Non-carbon overshoot: 0.756 − 1.0 < 0, so 0.000 — there is none. The whole of the reported overshoot sits inside the carbon term. The figure therefore supports no conclusion about crops, grazing, timber or fish; it supports a conclusion about atmospheric carbon, which is better supported by three direct physical measurements. Credit any working that reaches the non-carbon figure and states the limit of the inference.
10. The same £4,000,000 programme, the same 15 percent chance the load-bearing assumption is wrong, the same 60 percent loss if it is — but the validation unit reports to the programme sponsor, and you estimate its detection probability at 25 percent. Does the function pay for itself at a 3 percent budget? What is the cheapest fix?
Expected loss: 4,000,000 × 0.60 × 0.15 = £360,000. Expected loss avoided: 360,000 × 0.25 = £90,000. Budget: 4,000,000 × 0.03 = £120,000. Ratio: 90,000 / 120,000 = 0.75× — it does not pay. Break-even detection rate is 120,000 / 360,000 = 33.3 percent. The cheapest fix is not more money: it is moving the reporting line, because P(detect) is the term that independence controls and budget does not. The strongest answers note that raising the budget while leaving the reporting line alone makes the ratio worse, not better.
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.
- Which of your own economic beliefs could you not currently state in a form that could come back wrong? Write one of them as an F-claim now, with its kill condition.
- Recall the last time you changed your mind about something substantial. What arrived — a number, a person, an experience — and what does that tell you about how to reach yourself next time?
- Where in your work are you the author and the reader of your own register? Who could be the second reader, and what has stopped you asking them?
- Think of a criticism you received that you dismissed and later found was right. What did you dismiss it for — the content, or the way it arrived?
- What would you have to see to abandon the position this book has been arguing? Write it down before you read on. If nothing would, sit with that for a while; it is information about you rather than about the book.
- When did you last answer a specific empirical objection with a general philosophical one? What were you protecting?
- Which of your commitments has no kill condition because you would keep it anyway — a person, a practice, a place? Notice that this is fine, and notice that it is why such commitments cannot be defended as though they were evidence.
- If someone paid you well to attack your own best work, what would you go after first? Write the three lines. You have just done the valuable half of the job for free.
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 register a discipline or a defence? A textbook that publishes its own kill conditions may be doing something rigorous — or it may be inoculating itself, converting the whole of the critical literature into a list it controls the wording of. Argue either case. Engage Lakatos on degenerating programmes and at least one source on pre-registration or registered reports that the chapter does not cite.
2. Hayek's objection, taken seriously. The chapter concedes that every alternative indicator is a centrally specified metric with none of the error-correcting property of a price. Argue either that this is fatal to "beyond GDP" measurement, or that the concession is smaller than it looks because the relevant domain is precisely where prices are absent by construction. Use Hayek (1945) and Grossman and Stiglitz (1980), and at least one source on indicator gaming or Goodhart's law that the chapter does not cite.
3. The rarity problem. Hansmann's explanation — that governance cost, not capital hostility, keeps worker ownership rare — makes a prediction about where cooperatives will be found, and it is broadly right. Write the strongest case that this settles the matter against the cooperative form, then the strongest rebuttal from the survival and productivity evidence. Conclude with which you find more persuasive. Use Hansmann (1996) and Burdín (2014), and find one post-2015 empirical study the chapter does not cite.
4. Scale. Ostrom's design principles were derived largely from resources with fifteen to a few thousand users, clear boundaries and long acquaintance. Argue either that polycentric governance extends to the atmosphere, or that the extension is a category error and the atmospheric problem requires an instrument of a different kind. Use Ostrom (1990) and Araral (2014), and at least one source on international climate governance that the chapter does not cite.
5. Strike the title. The chapter's central concession is that nine of twenty-four load-bearing claims forbid nothing, that these nine are the ones the paradigm is named after, and that none of them is an input to any instrument in the edition. Argue either that this vindicates the edition — the operational core stands alone — or that it guts it, because a framework whose organising idea does no work is a toolkit with a story attached, and the story is doing something the author will not account for. Use Mirowski (1989) and the register itself, and one source on the role of metaphor in scientific explanation that the chapter does not cite.