Every poverty argument is really a fight over which of two definitions is in play: subsistence (the minimum to sustain life — essentially fixed across all human history) or participation (the minimum to function as a member of your society — Adam Smith’s linen shirt). The vault’s contribution is Chris’s ratchet objection: participation is a negotiated state, not a fact. If everyone lost internet access tomorrow, employers would revert to paper — so “X is required” is true only because X is common, and X becomes more common because it is required. That is a self-supporting feedback loop in which everyone eventually needs everything, and a standard built on it can never be satisfied by growth. It is the same structural defect as relative poverty being definitionally permanent, arrived at from the other side. But the participation view is not simply wrong, and the reason is a scope distinction: conventional necessities are collectively negotiable and individually binding. Which is exactly why the two camps never resolve — one is describing a constraint a person actually faces, the other a convention a society could change. Two further results the page carries: the standards do not differ by having social content (subsistence runs through social instrumental chains too — you need to show up at work) but by what the chain is anchored to — a physical fact that doesn’t move, versus what other people have, which moves whenever anyone acquires more; and the field trap that actually retires the subsistence standard is not an argument but a statistical swap — a minimum threshold compared against an average cost, which is a tautology wearing the clothes of a finding.
Links: Economics, Positional Goods (the material twin — that page is about goods, this one about needs), Poverty in America Is a Sign of Exploitation (prep) (the origin specimen — Chris argued subsistence, his opponents wanted participation), Opening Statement — Poverty and Exploitation, Human Constancy (what §6c actually requires — ends drift slower than means, not human nature is constant), Scope Confusion (the mechanism behind the impasse, §4), Reading Outcome Statistics §5 (the floor-is-not-a-mean diagnostic this page’s §6a supplied), Absolutes and Differentials, The Load-Bearing Word, Measuring Inflation, Civilizational Cycles — structural-institutional (Turchin; §6)
Thesis page (portable). Sparked by the Econ Nerds inequality review, but the live specimen predates it: this is the definitional fight Chris was actually in.
| Subsistence | Participation | |
|---|---|---|
| Poverty is | lacking the minimum to sustain life | lacking the minimum to function as a member of society |
| The bar is set by | biology | custom |
| Over time it | stays put | rises with the society |
| Can it reach zero? | Yes | No — see §3 |
| Champion | classical/absolute measures | Adam Smith |
The participation standard is not modern goalpost-moving, and anyone arguing subsistence should know this before they get ambushed with it. Wealth of Nations, Book V ch. 2:
“By necessaries I understand not only the commodities which are indispensably necessary for the support of life, but whatever the custom of the country renders it indecent for creditable people, even of the lowest order, to be without. A linen shirt, for example, is, strictly speaking, not a necessary of life… But in the present times, through the greater part of Europe, a creditable day-labourer would be ashamed to appear in public without a linen shirt.”
So this is a two-century-old dispute inside classical economics, and the founding text is on the participation side. Treating it as a recent ideological invention is a factual error and a losing move in a debate.
The Econ Nerds video walks the US poverty measures — official → Supplemental (SPM) → absolute-threshold — and lands the key observation aloud: “this is actually a relative measure because they keep increasing the poverty threshold.” Chris had made the same move independently in his own debate prep:
“Relative poverty is definitionally permanent — triple everyone’s real income and the rate doesn’t move. A metric that can never reach zero cannot be a sign of anything.”
Two independent derivations of the same point is the argument for its soundness. It is also the load-bearing word in its native habitat: whoever pins poverty has set the win condition, and the pin almost never gets contested because both sides assume their own reading is the obvious one.
The participation standard has a defect deeper than “the bar rises.” The bar rises because of itself.
Chris: “The trouble here is that ‘Participation’ is a negotiated state. If all of a sudden no one has internet, the market will adapt to not make this a requirement. So arguing ‘participation’ is the same as ‘required’ is a self-supporting feedback loop where everyone eventually needs everything.”
Unpacked:
The requirement is endogenous. Nothing about a job application intrinsically requires a broadband connection; the connection became necessary because enough people had one that the alternative could be withdrawn. Contrast calories, which are required by metabolism whatever anyone else is doing. That is the first cut at the real distinction — a physical requirement versus a coordination equilibrium — and it is not the distinction either side usually argues. (§3a sharpens it: the split is not whether social content is present, since subsistence has plenty, but what the chain is anchored to.)
The consequence is fatal for the participation standard as a measure:
Both relative poverty and participation poverty are defined by reference to the distribution. Growth moves the whole distribution. Therefore neither metric can ever register growth, and neither can ever reach zero.
A metric that cannot reach zero is not measuring a problem. It is measuring a shape. This is why the standard is attractive to an advocate: it makes the affirmative burden permanently satisfied.
Chris’s term is negotiated, and he means it in the economic sense — not people bargaining at a table but market forces settling on an equilibrium, the way they settle everything else. No single individual or firm sets the requirement; the market arrives at it, and can arrive somewhere else if conditions change. That is the sense in which the requirement is soft despite being real.
But he also supplies the objection to his own clean split, and it is the better version:
Chris: “and same to the physical requirements… you need to show up at work!”
Right — and this dents §3’s tidy line between physical and conventional. Subsistence needs run through instrumental chains that are themselves socially set: you need calories → you need income → you need to show up → you need transport. Transport is now instrumentally subsistence-required, and what counts as adequate transport is set by market and settlement patterns, not biology. So subsistence is not the pure category it looks like.
The repair, and it is what actually separates the two standards:
Both standards run through social instrumental chains. They differ in what the chain is anchored to. Subsistence’s chain terminates in a physical fact that does not move — you need the calories whatever anyone else is doing. Participation’s chain terminates in what other people have, which moves whenever anyone acquires more.
That is why the ratchet applies to one and not the other. Subsistence’s instrumental links can inflate — and do — but they are pulled taut against a fixed endpoint, so the standard cannot run away indefinitely. Participation has no such endpoint. It is the anchoring, not the presence of social content, that does the work — and conceding the social content up front is what stops the argument being lost on the “you need to show up at work” reply.
The ratchet does not make the participation view false, and the honest refinement is what explains the impasse:
Conventional necessities are collectively negotiable but individually binding.
Society could restore the paper application channel. No individual can. So at the level of the person, “I need internet to get a job” is simply true — a hard constraint, not a preference. At the level of the society, “internet is a necessity” is a convention that could be renegotiated. Both statements are correct at different scopes, and each side hears the other as obviously wrong because each is speaking at the scope where it is right.
That is scope confusion doing what it always does: producing an unresolvable argument out of two true claims. Naming the scope is the move that unsticks it.
Since “necessity” splits, here is the test that decides individual cases:
Did the cheap path get abolished, or did it merely become unfashionable?
| Claimed necessity | Was the alternative withdrawn? | Verdict |
|---|---|---|
| Internet / phone | Yes — paper channels closed, benefits and banking moved online | Imposed constraint |
| Car (most US metros) | Yes — zoning and land use removed the walkable option | Imposed constraint |
| A newer phone every two years | No | Want inflation |
| A large television | No | Want inflation |
This keeps the subsistence position honest without conceding the ratchet. Some claimed necessities really are imposed — and note that in both imposed cases above, the imposition was a policy choice, which routes straight back to manufactured positionality.
And a genuine concession to the subsistence side, which is Chris’s: “living like a 1963 person” may be far less punishing than it sounds. The goods a 1963 household actually consumed — food, clothing, appliances, entertainment — are radically cheaper now. The expensive part of modern life is shelter, care, and credentials, which is the positional basket, not the participation basket. The participation ratchet and the cost-of-living squeeze are largely about different goods, and conflating them is how the argument stays confused.
Credential inflation is this page’s mechanism applied to jobs. A role that took a high-school diploma in 1963 now takes a bachelor’s — not because the work changed but because the signal inflated. More people got degrees, so a degree stopped distinguishing anyone, so the bar rose. Steps 1–4 of §3 exactly.
This bears on Turchin’s elite overproduction, and suggests Chris’s read — that Turchin was being too precise — is right:
If the broader mechanism is the real one, that would explain the 2023 PLOS ONE result already recorded in the vault — elite overproduction underperformed inequality alone as a predictor of instability. A narrow operationalization measuring a subset of the mechanism should underperform. Being less precise might make the theory more predictive, which is an unusual and testable claim.
Related but distinct — do not merge them. Wage polarization (Autor) and elite overproduction (Turchin) are not competitors:
| Wage polarization | Elite overproduction | |
|---|---|---|
| Population | the deskilled middle | the overcredentialed top |
| Side | demand-side — automation destroys routine tasks | supply-side — too many aspirants per slot |
| Failure | can’t earn above poverty | can’t reach expected status |
Chris’s 1963-man point (“the 1963 man is not going to have a wage greater than the poverty level”) is the polarization story — a skills-and-automation claim about the middle, not Turchin’s claim about the top. Both can be true simultaneously; they explain different people.
The live danger is not that anyone refutes the subsistence standard. It is that it gets discarded as meaningless — “nobody cares about bare survival, what matters is participation” — and the substitution is made with evidence rather than argument. Chris met the exact move:
Chris: “In the debate this was challenged by giving me studies talking about the cost of living across states and showing how the current poverty levels were well below these. Upon examination, the studies were all using AVERAGE cost of living per state, and thus of course it was above poverty by definition :)”
A minimum threshold compared against an average cost cannot come out any other way. For any non-degenerate distribution the mean sits above the floor, so “the poverty line is below the cost of living” is a tautology wearing the clothes of a finding — true numbers, reputable sources, nothing established. A poverty line that were not below average cost of living would not be a poverty line.
It slips past because both quantities share a unit and a subject, so the swap never looks like a category error. Filed as the fifth diagnostic on Reading Outcome Statistics §5 — a floor is not a mean, where the general repair lives: before comparing two statistics, check they are the same kind of statistic.
Why this is the trap and not merely an error: it lets the participation standard be installed rather than defended. If the subsistence number can be made to look absurd, nobody has to argue that participation is the right standard — the field is simply conceded. Which is why §7’s first instruction is to pin the definition before any evidence is exchanged.
The deepest point in the discussion, and it dissolves most of the argument:
Chris: “‘Want’ is endless, but economics deals in what is possible from those wants, and the idea that now, more than ever, we can sustain more human life than ever before is a finding separate from how well people live.”
Two questions, routinely fused:
| Capacity | Distribution / participation | |
|---|---|---|
| Asks | how many humans can be sustained above the floor? | how well do people live, relative to each other and to expectation? |
| Answer trend | unambiguously, enormously up | contested, and by construction never settled |
| Refutes the other? | No | No |
The Econ Nerds video is championing the first. Its critics answer with the second. Neither refutes the other, and both are true — which is why the exchange never resolves and why each side reads the other as morally obtuse.
And the anchor underneath it, which is why capacity is the achievement:
Chris: “The default state of all men IS poverty. Getting humanity above this is the big achievement.”
This is the poverty-as-baseline argument in its general form: everyone starts at zero, so poverty is the origin, not a deviation requiring explanation — what needs explaining is wealth. It also supplies the sharpest reason the participation standard cannot be the only standard: a measure defined against prevailing consumption cannot register the one achievement that actually happened, because it re-baselines on it. Escaping the default state is invisible to a metric anchored to the new default.
The honest counter-pressure, kept on the record: none of this establishes that capacity is the only question worth asking, and “the pie grew” is not an answer to someone whose complaint is about the positional basket. Two findings means two, not one that outranks the other.
The open question was whether any “necessity” has ever been de-necessitated. Chris’s answer reframes it, and the reframing is better than the question:
Chris: “We often find ourselves in a situation where early forms of technology are not polished enough for general use. You get lots of specialists here and things get overly complicated. But then the tech improves, it becomes well adopted, and commonplace… I have seen people who take the bus to work with the latest phones. And this is not just tech. We often see the importance of family and religion come back DUE to the complexity a modern society brings. I think the point here is that humans still ‘human’ even after all this time… In many ways we are more alike our caveman brethren than not.”
The ratchet runs on means, not on ends. What inflates is the set of goods required to satisfy a need. The need itself does not inflate — it is a fact about the animal, and it has not moved since the cave. So:
So the honest answer to “can it run backwards” is: the ratchet does not reverse, but it is bounded — because it is satisfying a fixed set of human ends. A means that stops serving its end gets abandoned however entrenched it is (cheques, landlines, cursive, the fax), and a need that modernity stops meeting comes back looking for a new means. The bound is human nature, not metabolism — which extends §3a: subsistence’s chain is anchored to a physical fact, and the wider participation chain is anchored to a psychological one that turns out to be nearly as stable.
What this actually requires — the assumption, downgraded. This section was previously flagged as resting on “human nature is constant,” its weakest load-bearing claim. That overstated the requirement. The ratchet does not need constancy; it needs only that ends drift far more slowly than means — a much smaller target, and one with evidence behind it. See Human Constancy, which supplies the instrument: across a 3,000–4,000 year gap the ends-content of ancient texts is transparent while the means-content is opaque, and both crossed the same gap through the same translators and the same survival filter. If ends drifted at the rate means do, both would be equally alien.
Where it could still be wrong: the falsifier is a durable change in what humans want, not in how they get it — and fertility collapse is the case that will decide it, since reproduction is the paradigm fixed end and people across every developed society are choosing dramatically less of it. Either the end moved, or an unchanged end is being defeated by its means-environment. The means-substitution reading is available and not established; that page keeps the objection live rather than resolving it in the thesis’s favour.
The parked open question was whether Sen’s capability approach supplies a participation-style standard that does not ratchet. Chris’s response:
Chris: “The market dictates what is required and adapts to circumstances… as soon as a ‘baseline’ is made, it becomes expected and assumed, and someone is now looking to increase that baseline. The market is the check here.”
State plainly what this is: an anticipated counter, not an engagement. Nobody here has read Sen properly and the page should not pretend otherwise. But the anticipated counter has a sharp form worth recording, because it locates where the fight will be.
Sen’s move is to anchor the standard to capabilities and functionings — being nourished, being sheltered, being able to appear in public without shame, participating in the life of the community — rather than to the commodities that deliver them. That is a genuine answer to the ratchet, because commodities inflate while functionings are meant to be stable. It is §6c’s means/ends split turned into a measurement framework, which is a real achievement and should be conceded as one.
The problem is that the capability list is heterogeneous. “Being adequately nourished” is anchored. “Being able to appear in public without shame” is Adam Smith’s linen shirt — definitionally the ratcheting one, since shame is measured against what others have. So the capability approach does not dissolve the problem uniformly; it relocates it to the question of which capabilities are on the list, which is the Weighting Problem wearing new vocabulary.
That is the argument to make against Sen if it survives contact — and reading him properly to find out remains the biggest live gap on this page.
economics · philosophy · politics · debates