This book is about how strongly other people's behaviour shapes our own, and why this is key for public policy. The author moves clearly and steadily from core research findings in behavioural lab and field experiments to historical episodes where societies reach then tipping points when a given behaviour flips rapidly from frowned upon to widely accepted, or the converse (think indoor smoking or sexual revolution). The key of these tipping points lies in behavioural contagion: as long as we don't observe a given behaviour, we tend to think that it is scarce, even if it is in fact widespread but concealed. Once we recognize the true extent of the behaviour, we change our own (and our acceptance of it) accordingly. At the level of a society, such changes can be extremely fast.

This, the author argues, should be a prime topic for public policy. As he repeatedly argues, there is solid evidence that being in contact with more smokers increases significantly one's odds to start smoking (and regretting it). The damage thus caused by the sheer presence of other smokers actually dwarfs the damage due to second-hand smoke. The book documents several other issues of this kind before culminating with what the author calls *the mother of all cognitive illusions*.

In a nutshell, it is the idea that higher taxes on the affluent will make them worse off. In the face of it, it is a no-brainer: more taxes means less money to spend on what I want. The trick lies in the fact that above a certain threshold (that a majority of households do reach in rich countries), we are driven willy-nilly to consuming position goods, that is good that do not provide us with much higher benefits than cheaper ones, but do improve our standings or prospects relative to others. Housing is a prime example. McMansions from the top lead people across a broad range of revenue to aspire to larger dwellings (which are more expensive to maintain and less environmentally friendly). It is often difficult to escape this: even if you recognize that some neighbourhoods are way overprices, you may still want to buy a house there because schools are better. So do other households, leading to a damaging bidding war that profits only to house sellers. In this example, a tax on house prices whose proceeds would go to funding better schools everywhere would have the effect of leaving everyone better off: it would cool down bidding wars, and decrease price differences between neighbourhoods. At a society level, a steeply progressive and comprehensive income tax would have the same effect: the richest would still drive luxury cars, a bit less overpriced, and there would be much more money for infrastructures and public services, leaving everyone better off.

The core argument is not new in itself: people have long pointed at how wasteful keeping up with the Joneses can be(ancient writers already nailed it). What is interesting here is that the author does not link this behaviour with some fundamental flaw of human nature, such as greed or envy, but on rational motives: we are influenced by the behaviour of others because of build-in reactions that allowed our species to survive without claws, furs or fast legs. Even in modern societies, such behaviours are not bugs, they are features that enable social interactions.

Compared to other popular books in the field, starting with *Nudge*, this book is a change of pace. Where most books deal with narrowly-defined issues and carefully crafted experiments, this one deals with society-wide issues, and equally far-ranging policies. It thus provides an answer to the (in my opinion unfair) idea that behavioural insights are limited to small-scale issues and unfit for the big challenges of our age. ( )