In this book, the author applies rational choice theory (and game theory to a lower extent) to explore the hidden side of the economics/mechanisms in various fields, from gambling to sex, politics, urbanism or economic growth, with the goal to question conventional wisdom and unveil counter-intuitive forces at play.
Using research and conclusions achieved by others, Tim Harford strives to demonstrate that rational behaviour is more widespread than expected. Key to his reasoning is the role of incentives (may they be of a financial or emotional kind) to explain people's individual and collective behaviour.
I find that this kind of story-telling compilation book has therefore the benefits of fluidity and diversity in its themes, but also the drawback of a lack of structure or comprehensiveness.
Various valuable concepts are illustrated, and the following ones in particular.
Rational individual behaviour does not necessarily lead to a desirable outcome for the majority (or eventually for the individual itself), because of the different intensity of each stakeholder's interest, and therefore of the actions they take, because of the dilution effect, or because of the external forces at play. Domains of application range from top executive pay to lobbies.
Economics can sometimes supersede psychology, as it is the case with Giffen goods for instance, which are such a necessity that when the price rises, demand rises too (although not linked to negative price elasticity).
People value objects higher when they own them: the endowment effect (ex: someone who owns a wine bottle which gained value over time won't sell it afterwards but drink it, although he would not buy such an expensive bottle).
In Thomas Shelling's model of human interactions, a negotiation is not only optimised by mathematics (Von Neumann's model) but by "focal points" that are invisible under a mathematical formulation (ex: 10% pay rise bargained by a trade union: 10% may not have a mathematical justification, but it is used as a point of reference). From this perspective, game theory works best when used in conjunction with simple strategies using focal points (ex: arbitrary references, taboos), for nuclear dissuasion for instance. What also emerged was the idea to use a local difficulty as a focal point to co-ordinate efforts to achieve a more broad-scale result.
The concept of little imbalances leading to an extreme state of segregation can be modelled by Shelling's chessboard simulation, with an application to urbanism.
The human factor is at play when a large group of people in cities, in companies’ clusters or inside a large company interact, which results in knowledge spillovers. The larger the group, the more likely it is for someone to come up with a new idea, which would then benefit everyone in the group, and would generate new ideas via a positive feedback loop. From a global perspective, Michael Kremer observed that the rate of technological progress is proportional to the world's population.
A key success factor for the human race (Homo Sapiens) over the Neanderthals (which they replaced) was the division of labour, which in turn led to trading, seen by Adam Smith as a quality of humans over animals.