Robin Hanson on Signaling and Self-Deception

Robin Hanson — economist at George Mason University and co-author of The Elephant in the Brain — joins Tyler Cowen for Ep. 35 to argue that hidden motives, not stated ones, drive the vast bulk of human behaviour, and to explore the implications for policy, governance, and personal identity.

Key ideas

1. Signalling explains well over 90 per cent of behaviour in a rich society. Hanson’s thesis — developed with Kevin Simler in The Elephant in the Brain — is that the motives people give for their actions (learning, health, helping others) are systematically less true than the underlying motives (showing off, signalling group loyalty, demonstrating care). The gap is not cynicism but a structural feature of how social creatures coordinate.
2. Self-deception is the engine, not a side effect. The mechanism is not conscious lying but genuine unawareness: we are built to pursue hidden motives while sincerely believing the stated ones. Policy analysts and social scientists who miss this will design institutions around the wrong targets.
3. The real remedy is a more informed audience, not less signalling. Taxing or restricting signalling is practically impossible because nearly everything we do has a signalling component. The more tractable lever is raising audience sophistication: when the people we are trying to impress know what actually works — which medicines heal, which charities help — we are forced to do things that actually work.
4. Prediction markets reveal the hidden motives of organisations. Corporations resist prediction markets not because they distrust information but because they rely on the excuse that forecasting failure was unforeseeable. A market that tracks project-deadline probability from the outset destroys that excuse and threatens the political coalitions that depend on it.
5. Futarchy — governing by prediction markets — has at least a 30 per cent chance of significant adoption within a century. The core idea is to vote on values and bet on beliefs: measure national welfare, let markets forecast the effect of each policy on that measure, and adopt whichever policy the market says will score highest. The obstacle is not logic but the multiple equilibria of social norms — prediction markets face the same adoption hurdle as cost accounting once did.

Content

The elephant in the brain: hidden motives as the default

Signalling (in the economic sense used here) means using a costly or hard-to-fake action to communicate something about yourself — think of a peacock’s tail, or a charitable donation made publicly. Hanson extends the idea far beyond the obvious cases. His claim in The Elephant in the Brain is that medicine, education, charity, art, conversation, and most other major social institutions are organised around displayed motives — health, learning, generosity, culture, connection — that are systematically less true than the underlying ones, which are almost always some combination of showing ability, showing loyalty, and being seen to care.

When Cowen presses him on what percentage of behaviour this covers, Hanson answers simply: ‘In a rich society like ours, well over 90 percent.’ The qualifier matters — he is talking about the proximate social explanation for voluntary behaviour, not the ultimate physical or evolutionary one. He is also careful not to call this hypocrisy in a moralistic sense: humans are built this way, self-deception is part of the design, and what they actually are — not what they pretend to be — is, he says, ‘spectacular.‘

Self-deception and why social nerds have an edge

The mechanism Hanson describes is not strategic deception but genuine unawareness. We are not consciously gaming our stated motives; we sincerely believe them, which is what makes them effective socially. This has an unexpected implication for who is best placed to study social behaviour. Socially skilled people navigate the world by intuition and never notice the gap between their textbook theories and their actual conduct. Socially awkward people — what Hanson calls, with good humour, ‘nerds’ — are constantly puzzled by the gap, and that puzzlement drives them to theorise explicitly. Being bad at a skill, he suggests, can be an advantage in analysing it.

Why you cannot tax signalling — but you can improve the audience

If 90 per cent of behaviour is signalling, taxing nonsignalling seems logical — but Hanson dismisses it quickly. Nearly everything has a signalling component; you would end up paying people to sit alone, and even that they could turn into a signal of depth. The more actionable prescription is improving the sophistication of whoever we are trying to impress. His example is medicine: we spend heavily on healthcare to signal that we care about the people around us, but because neither patient nor audience knows which treatments actually work, we do not optimise for treatments that work. A more informed audience would pressure us to care more about outcomes. This is, Hanson notes, an odd route back to Matthew Arnold’s elitism — not because elites are intrinsically better, but because a knowledgeable audience sets a higher bar for the signals it rewards.

Prediction markets: why organisations resist them and why they matter

Hanson’s career as an institutional designer is centred on prediction markets — mechanisms where people bet on outcomes and the market price aggregates information about the likely result. His account of why corporations resist them is a case study in his own thesis. Managers claim they want more information and analysis — ‘collecting information’ is the universal excuse for almost any activity — but what they actually want is the ability to blame unforeseeable events for failures that were foreseeable all along. A prediction market on a project deadline, visible from the start, makes it impossible to claim nobody saw it coming. It threatens the political excuse structure on which organisational careers depend.

He is similarly wry about the enthusiasm prediction markets have generated among the rationality community and in books like The Wisdom of Crowds and Superforecasting. The popular interest, he argues, is less about building better institutions than about using prediction markets as a status game — identifying who among us is cleverer and more calibrated, rather than creating prices that would benefit everyone.

Futarchy: betting on beliefs, voting on values

Hanson’s proposal for prediction markets at the scale of governance is called futarchy. The idea has two steps: define a measure of societal welfare (GDP per capita, a composite wellbeing index, whatever voters choose); then, for each proposed policy, open a conditional prediction market that bets on what the welfare measure would be if that policy were adopted versus not. Whichever policy the market predicts will score higher gets implemented. ‘Vote on values, bet on beliefs’ is his summary.

He is honest about the adoption timeline. In a world where nobody uses prediction markets, proposing them sounds like an accusation — ‘people are bullshitting around here.’ In a world where everyone uses them, refusing them sounds equally suspicious. He sees the challenge as switching equilibria, much as cost accounting once had to, and puts at least a 30 per cent probability on substantial futarchy adoption within a century, conditional on the Age of Em not intervening.

The Age of Em: brain emulations and Fermi’s paradox

Hanson’s earlier book, The Age of Em, models a hypothetical future in which brain emulations — digital copies of human minds, running in virtual reality on real computing hardware — become cheap enough to take over the global economy. He treats this not as speculation but as rigorous extrapolation: if brain emulation becomes feasible, the resulting economy would grow so fast, and consume so many physical resources, that it would quickly become detectable on a cosmic scale. This connects to the Fermi paradox — why do we see no such signatures from other civilisations? His preferred answer is that the great filter (whatever makes matter fail to produce visible civilisations) is most likely behind us: the very first step toward life was the hard part, and if that is so, our future could be bright. He holds this view with appropriate uncertainty — a 20 per cent residual probability that the filter lies ahead is not, he insists, something to rest easy on.

On rational thinking, specialisation, and the rationality community

Cowen asks what would most improve collective rational thinking. Hanson answers: more prediction markets, and alongside them, more specialisation. The instinct to have opinions on everything — which a format like this conversation actively encourages — is, he says, one of the clearest failure modes of public discourse and financial markets alike. You will be a better thinker, and a better trader, if you know some things well and admit confidently that you do not know the rest.

His critique of the rationality community follows the same logic. Groups that coalesce around the goal of being rational tend, in practice, to collect signals of rationality — knowing Bayes’s theorem, being able to cite the replication crisis — rather than actually improving their beliefs. The group’s ability to verify rationality is too low; the temptation to crow about membership is too high. It becomes yet another arena for the elephant in the brain.

Related

• Robin Hanson — speaker; economist and co-author of The Elephant in the Brain
• Rationality — concept; Hanson’s critique of the rationality community engages this page directly
• Many-Worlds Interpretation — concept; Hanson discusses the quantum-mechanical many-worlds view and personal identity