Currencies of Value

What are you counting?

An advertiser spends forty thousand dollars on a campaign and buys, in return, several million impressions. Nobody blinks at this transaction. Money in, attention out — a clean exchange rate, quoted daily, adjusted in real time by auction. Attention has a price, and money is how you pay it.

This works so smoothly that it's easy to miss what's actually being claimed: that two genuinely finite resources — money and attention — are being traded against each other, and that the trade is happening on money's terms. Your Attention Belongs to Someone, earlier in this series, made the case that attention is scarce in exactly the way money is: it can't be manufactured, only redirected from wherever it currently sits. The question that essay left open is whether people notice they're spending it. This essay starts from a related question: once you admit attention has a price, what else turns out to have one, and does the same kind of trade always work the same way?

Salience told you attention was misallocated. It didn't tell you against what.

The Priorities and Salience essay in this series named salience as the gap between how much a problem matters and how much notice it gets — soil degradation losing to hurricanes not because it matters less, but because it's slower and less photogenic. Seth Godin lists attention criteria that go further, naming the specific traits (visibility, urgency, status, controllability) that determine what gets noticed almost independent of severity.

The salience essay and Godin's criteria both diagnose a mismatch. Neither one specifies the unit the mismatch is measured in. "Attention is disproportionate to severity" is a comparison — and any comparison needs two things measured the same way to be effective. How is severity denominated? Lives? Dollars? Something else? The salience argument borrows a sense of "how much this matters" without ever cashing it out. It leaves a question sitting underneath the whole framework, waiting to be asked directly.

So: what are the actual units on offer, and what happens when you use them as rival measuring instruments rather than interchangeable stand-ins for "importance"?

Three categories, and one refusal to fit

Money, attention, users and followers, lives, virtue — all candidates people reach for when asked what matters most. They don't all do the same job. This is demonstrated by sorting them by function rather than popularity:

Stakes currencies try to measure what's actually at risk. Lives — and the machinery built to sum and compare them, the Quality-Adjusted Life Year (QALY, a year of life weighted by the health quality lived in it) and the Disability-Adjusted Life Year (DALY, a year lost to premature death or disability) — is the clearest example: it denominates suffering and death directly, on the premise that these can be counted and added across a population. Cost-if-unsolved, described in Priorities and Salience, is a related but distinct instrument: not a headcount, but a measure of how close a problem sits to a point of no return.

Reach currencies measure something different: not what's at stake, but how far notice travels. Here a real distinction hides inside what looks like one category. Attention can be framed as an individual, experiential reading — what one person notices, moment to moment, before conscious evaluation even starts. Monthly active users (MAU), the metric many businesses actually report to investors, is the aggregate, institutional version of the same underlying phenomenon — a platform's retrospective count of how many separate individual-attention events it captured over a month. These aren't two different things competing for the same job. They're the same axis read at two different scales — one from inside a single mind, one from a dashboard. They're worth keeping distinct precisely because conflating them can be misleading: a story that goes viral (high aggregate reach) hasn't necessarily earned anyone's sustained individual attention, and something one person attends to closely for years may never register at the aggregate scale at all. Social capital — the favours owed, the calls that get returned, who can get whom on the phone — is a third variant of the same axis: reach weighted by relationship strength rather than by headcount or moment.

Credit currencies are different in kind from the first two. They don't measure a problem's stakes or its reach. They measure whether an actor deserves esteem for how they behaved — a verdict on conduct, not a quantity of anything. Two distinct currencies live here, close kin but not the same thing. Virtue is a verdict on disposition: did you act well. Trust is a verdict on reliability: will you act as expected, does your word hold. The two usually travel together but can come apart cleanly: a salesperson who's bluntly honest about a product's flaws earns trust without necessarily being virtuous in any fuller sense, and someone can be judged virtuous in the abstract while still not being trusted with a specific task they haven't yet proven themselves on. Virtue is closer to a disposition; trust is closer to a track record. Rutger Bregman's Moral Ambition is built around trying to make virtue function as a scored currency. Both virtue and trust resist the other currencies' logic in a way worth analysing rather than treating as a measurement inconvenience to be solved with better instruments.

Money doesn't sit inside any of these three boxes, and that characteristic is the interesting thing about it. It can buy reach (the advertiser above). It can buy the trappings of stakes-worthiness (a cause dressed up as urgent). It can even try to buy credit. It's tempting to read this as money being the one true currency underneath the other three — Yuval Noah Harari's account of money as humanity's most successful shared fiction argues that money's power comes specifically from being the only value-system universally convertible into any other. But convertibility isn't the same as equivalence. Money converts cleanly into reach because reach, like money, is at bottom just a count of a scarce thing moving. It does not convert into virtue — only into virtue's appearance, which is a different and lesser thing. Money is the currency best at acquiring the other three. That's not the same as being the one true currency underneath them.

What money can and can't buy

A useful frame, used in career-coaching circles, is that a person's effort pays out in five distinct "salaries," not one: financial (the money itself), psychological (meaning felt in the work), social (status and identity capital), educational (skill and mastery banked for later), and freedom (autonomy to work, or not, on your own terms). This essay treats educational salary specifically as skill and mastery, not the broader "relationships" the term sometimes covers — relationships are social salary under another name. The remaining distinction, psychological versus educational, deserves a beat: one is a stock, meaning felt right now; the other a flow, capability banked for use later. The apprenticeship model trades almost entirely on the second — low pay, low status, low immediate satisfaction, accepted because the capability being built is worth more later than any other payout would be worth now.

Money's link to financial salary is close to tautological. The interesting case is what happens when money is spent toward the other four — and here it forks in a way worth naming directly. Spent on a bigger house or a more visible title, money buys social salary: status, what economists call a positional good, valued for how it ranks you rather than what it does. Spent on savings, or the ability to say no to a client, the same money buys freedom salary instead — and the two are not just different, they compete: money spent chasing status is money that didn't go toward the optionality that freedom salary is built from. Two people on identical incomes can be doing entirely opposite things with them. Money looks like a single currency because it has one unit. It behaves like two, depending on which salary it's converted into.

Virtue: the currency that resists being made into one

Bregman's central move in Moral Ambition is to try to make moral contribution function as a competitor to conventional status — to redirect what confers social salary away from wealth and toward genuinely doing good, so that the same status-seeking instinct that currently drives consumption and competition gets pointed somewhere better instead. The mechanism behind this is real and well-evidenced: Robb Willer's research on collective action found that people who contribute more to a group don't just feel good about it, they measurably earn higher status, more interpersonal influence, more cooperation from others — and that earned status then predicts still greater future contribution. Change what confers status, in other words, and you change what people actually do, not just what they claim to value. That's not a hopeful analogy; it's an experimentally demonstrated feedback loop.

But something happens the moment virtue is asked to function as a scored, comparable, competitive currency the way status conventionally does. It's a structural tension, and sociologists of quantification have a name for why.

Wendy Espeland and Mitchell Stevens' work on commensuration — the process of making different qualities measurable on one shared scale — makes the mechanism explicit. Commensuration doesn't discover a common unit that was quietly there all along. It manufactures one, and manufacturing costs something: incommensurables, the things that resist a shared metric, tend to show up precisely at the boundary between institutional domains where identity and values are actually in question. Virtue sitting at the boundary between the ethical domain and the status economy isn't a measurement failure waiting on better instruments. It's what a genuine boundary between two different kinds of value looks like from the inside. Force a single unit across it — a moral-ambition leaderboard, a quantified virtue score — and the unit doesn't reveal virtue's true worth. It erases the distinction that made virtue a different kind of currency from status in the first place. Trust resists the same move for a related reason: a trust score is already a slightly uncomfortable phrase, and the discomfort is informative — the moment reliability becomes something optimised for a number rather than demonstrated through actual conduct, what's being produced is no longer quite trust.

This is the deeper reason virtue and trust struggle to compete with money, reach, or stakes for attention in ordinary conversation. It isn't that anyone conspires to undervalue them. It's that the other currencies come pre-fitted with a scoreboard — a dollar figure, a user count, a life saved — and things without a scoreboard are structurally harder to point to, compare, or rally behind, whatever their actual weight. Virtue and trust don't lose the comparison. They're not entered into it in the first place.

The unit was never given. It was always chosen.

None of this — stakes, reach, credit, or money's refusal to sit inside any of them — exists because human beings are wired to measure value one particular way. Each is closer to a piece of technology than a fact about human nature, built by a specific set of institutions to solve a specific coordination problem, and replaceable when the institutions change.

QALYs and DALYs are a mid-twentieth-century artifact of health economics and utilitarian philosophy, built specifically to make triage decisions comparable across radically different conditions — a tool invented to solve a coordination problem, not a timeless unit that was always there waiting to be discovered. Attention-as-currency in its current form is inseparable from broadcast and social media technology; a famine needed to be narrated in text before photography, needed an image after it, needs a shareable clip under a minute now — the criteria for what counts as attention-worthy shift with the medium, not with the underlying severity of the famine. Money-as-universal-solvent requires deep capital markets to function as a pricing mechanism at all; a subsistence economy without reinsurance can't express "cost of inaction" in dollars because the market that would price it doesn't exist yet.

And status itself, per Willer's finding, is genuinely reconfigurable — not a fixed hierarchy waiting to be climbed by whichever means, but a resource that flows toward whatever behaviour a group currently rewards, and that reward criterion is a design choice, made and remade, not a constant. That's the empirical foundation underneath Bregman's whole project: if status attached to something else, people would still chase it, only now the chasing would point somewhere useful. The chasing was never the fixed part. The target was.

One more thing worth noticing before closing: several of these currencies carry a hidden clock, and the clocks don't all run the same way. A life-year, once spent, cannot be earned back — which is exactly what makes QALYs a stakes currency and not a reach one. Attention and MAU are also time-bound, but recoverably so: an ignored post can be resurfaced, a lapsed user can return. Money is the odd one out entirely — spent, it can be replenished by more of the same. The clock a currency runs on says as much about what kind of currency it is as the category it started in.

What this changes

Not the underlying stakes. A tipping point that's actually approaching doesn't move because someone picks a different currency to describe it in. What changes is smaller, and it's a decision rather than a discovery: which currency you're using for a given claim, which salary you're actually optimising for underneath it, and which of these a whole society has defaulted to treating as important. All three are choices, not facts about how value works — and the bookkeeping itself has limits. Some things, scored, stop being what made them worth scoring in the first place.

The advertiser buying attention with money is a transaction working exactly as designed — which is its own problem: a trade most people never notice they're making. Virtue and trust are a different kind of case entirely, and not because anyone is trying to game them. They simply don't come with a scoreboard, so they don't compete for notice the way a dollar figure or a user count does — not a flaw to be engineered away, but the consequence of measuring something that was never built to be measured that way. Adam Mastroianni describes moral behaviour as something we should expect of ourselves and encourage in others — not something that can be scored or bought. Expected, and encouraged.

This essay is part of a series on tools, cycles, and the problems worth working on.

Written with AI. Version 1 — September 2026.