No One Will Read the Label

The Economists of Proof — Herbert Simon

A supermarket, Tuesday, 6:40 pm. In front of the olive-oil shelf, a woman. Thirty options. On each bottle, a dozen mentions: origin, first cold pressing, labels, competition medals, various commitments. Roughly three hundred pieces of information for one litre of oil.

Observed decision time: four seconds. She grabs the bottle with the green label she thinks she recognises, and moves on to the pasta aisle.

Did she do it wrong? No. She did the only thing humanly possible — and understanding why is understanding what a label is really for. And why most of them no longer do it.

The man who demolished economic man

Economics reasoned for a long time about a fictional character: homo economicus, who compares every option, weighs every criterion, and chooses the optimum. Herbert Simon demolished him from within — patiently, over forty years, with a Nobel Prize at the end of it — by asking a question of disarming simplicity: with how much time?

Because human rationality is bounded. Not by stupidity — by construction: finite time, finite attention, finite computing capacity. Facing thirty olive oils and three hundred mentions, genuine optimisation would demand hours of reading, comparing, checking. No one has those hours, and no one ever will: it is the budget of the species, not of a rushed generation.

So we do something else, says Simon, and we all do it: we do not maximise, we satisfy — his word, which has stayed famous: satisficing. We take the first option that clears our threshold of “good enough,” and move on. This is not a flaw that a good consumer-education campaign would fix. It is the human condition — and any design that ignores it fails.

The scarce resource

In 1971, Simon wrote a sentence that describes our era fifty years ahead of time: what information consumes is obvious — it consumes the attention of those who receive it. Hence the reversal: a wealth of information creates a poverty of attention.

For the entire food sector, the consequence is brutal to admit: the eater’s problem was never a lack of information. There was already too much of it in the 1970s; there is a thousand times too much of it today. The real deficit is on the other side: four seconds of attention for three hundred mentions.

This is exactly why the label was invented — and it deserves to be said with respect: it is a rational invention. A label is a shortcut: the compression of everything one does not have time to verify — a specification, controls, a supply chain — into a sign recognised in one second. The previous episode in this series showed that price compresses scarcity; the label compresses quality. Two summaries for two things no one can know in full. The shortcut is not laziness. It is the only possible answer to the scarcity of attention.

The broken shortcut

But the two compressions do not share the same fate, because they are not disciplined the same way. The price is tested constantly: every transaction calls it back into question, every trade-off corrects it, it cannot lie for long. The label is disciplined by nothing comparable: between two widely spaced audits, it lives on its reputation — and its reputation lives on the memory of what it once required.

Hence the inflation. Since a recognised sign is worth gold, every actor has an incentive to create their own: public label, private label, house brand, in-house pledge, trade-fair medal. And every new sign taxes the attention the others were already sharing. Thirty signs on a shelf amounts to zero signs: the shortcut that was supposed to save attention ends up consuming it.

So the mechanism from the first episode starts up again, one notch higher. When the woman at the shelf can no longer tell a demanding label from a decorative one, she treats them all the same — at the average level. The label with the fierce specification is worth, within her four seconds, no more than the label printed for marketing. The demanding sign pays for the others, exactly as the good used car paid for the lemons. The market for shortcuts sinks the way the market for products sank — and for the same reason: nothing sorts the signs themselves any more.

Legible, not read

Here the reader is entitled to ask the awkward question: if proof is a path, and if Simon shows that no one will ever walk that path in front of a shelf — what is the point of building it?

The answer lies in a distinction the whole transparency debate conflates: being read and being legible are not the same function.

Look at companies’ audited accounts. No one reads them — not the customer, not the employee, hardly any shareholder. Are they useless? Quite the opposite: their power comes not from being read, it comes from being open. Anyone could pore over them. A few — an analyst, a journalist, a regulator, a competitor — sometimes do. And that mere possibility disciplines everyone upstream: accounts are not kept the same way depending on whether they are sealed or open.

Proof does not need to be read to act. It needs to be legible.

The shortcut at the shelf therefore remains indispensable — Simon demands it, and nothing will replace it. But there are two kinds of shortcuts: the one that leads nowhere, and the one that can be decompressed at any moment. The second disciplines what it summarises, precisely because someone can always go and check. The first summarises what no one watches — and it always ends up summarising less and less.

The woman at the shelf will never decompress anything, and that is entirely as it should be. Others will do it for her: an institutional-catering buyer whose liability is on the line, an inspector doing their job, a competitor looking for the flaw, a journalist on the day of a scandal. Four seconds of trust, backed by a complete path that others walk — that is a shortcut that works.

Back to the shelf, one last time. The goal was never a consumer who reads everything — Simon settled that question fifty years ago: that consumer will never exist. And that is good news, because a society where everyone had to verify everything themselves would be unlivable; trust by shortcut is a civilisational achievement, not a weakness.

The goal is a world where the four seconds are enough — because what they catch holds up to something. Attention is too scarce to be wasted on verification. All the more reason for what is not verified to be verifiable.

—

Herbert Simon, Nobel Prize in Economics 1978 for his work on bounded rationality — and Turing Award 1975: the only man to have stamped his name on both economics and computer science. The 1971 quotation comes from “Designing Organizations for an Information-Rich World.” Episode 7 of the series “The Economists of Proof.”