It would be so convenient if fundamental laws of nature told us how best to run a society. Governance would be a simple optimization problem, like finding the shortest route through a network; we could do without left–right political confrontation, and just solve the equations. Unfortunately, governance is not a well-posed problem. There must inevitably be balance and compromise: for example, of the rights of the individual against the overall good for society. This is what makes politics and economics not just controversial, but interesting.

Inequality is one of the biggest items on the agendas of both of these disciplines. Few people are likely to speak in favour of inequality as such, but in stereotypical terms the political right defends wealth as a reward for hard work, whereas the left deplores a society in which, as economist Joseph Stiglitz has said of the United States, “1 percent of the people take nearly a quarter of the nation’s income”. It seems an un­avoidable truth that a free-market capitalist system will create wealth inequality; to a free-market fundamentalist who sees markets as meritocratic optimizers of efficiency and resource utilization, that is not only necessary but moral. Under that philosophy, by intervening in the market in the hope of making the outcome ‘fairer’, we only throw a spanner in the works.

Yet even if one accepts some inequality as a necessary evil, there are options beyond laissez-faire. How, and how strenuously, governments and legislators should attempt to limit the extent of wealth inequality — crudely measured by the Gini coefficient, which quantifies the statistical dispersion of income distribution — is currently a hotly disputed matter. Should companies and banks be restricted in what they can pay their chief executives? Should taxes aim to inhibit or reduce the perpetuation of inherited wealth? Or is all this crypto-communist social engineering?

The strongest argument for such measures is not that it makes things more ‘fair’ (although meritocratic defences of free-market inequalities should surely at least demand a level playing field). Rather, it is that gross wealth inequality is socially corrosive. It polarizes attitudes, foments unrest (see, for example, the Occupy movement) and degrades trust and cooperation. At face value, a study published online this week in Nature supports that view — but with an added twist.

In the study, groups of volunteers played a simple economic game involving cooperation (a “public goods game”), in which they could lose or gain wealth through voluntary redistribution within social networks that started with three different levels of inequality (A. Nishi et al. Nature http://dx.doi.org/10.1038/nature15392; 2015). Crucially, in some games the wealth of participants was made visible to others, whereas in others it was kept hidden. For “invisible” wealth conditions, the games tended to converge on a fairly low Gini coefficient, but “visible” wealth produced higher (and less stable) average Gini coefficients. This result was exacerbated when the initial inequality was greater. In other words, simply hiding wealth decreased the wealth disparity in otherwise identical games and networks.

Still more importantly, visible wealth reduced the overall cooperation and interconnectedness of the social network, and in fact led to lower total wealth. As the authors say: “it is not inequality per se that is so problematic, but rather visibility” of that inequality. This fits with the established idea that it is relative, not absolute, differences in wealth that compromise happiness and promote discord: we resent what our neighbours have and we don’t. What grates is not knowing that others have more than us, but seeing that difference ostentatiously displayed.

It is dangerous, however, to think that these laboratory experiments can be extrapolated into a political or moral message for the real world. They invite us to frown on bling and the champagne-drenched excesses of financiers, but we should be cautious about their implications, even (or especially) if they flatter our preconceptions. Besides, there is scope here for upsetting both ends of the political spectrum. Right-wingers might deplore an injunction to hide one’s wealth, compromising personal freedom — isn’t it up to us how we spend our money? Left-wingers might dislike the idea of being relaxed about inequality as long as it is kept out of sight — and, anyway, might that not provoke a climate of secrecy and suspicion?

For now, the results should simply inform and broaden the discussion. They show, for example, that inequality is not solely down to market mechanisms, but also responds in subtle ways to our own dispositions. Above all, the findings are a reminder, along with related behavioural experiments on the role of punishment in public-goods games, that John Maynard Keynes’s “animal spirits” are an irreducible part of what shapes a market economy. It is time to lay the idea of the rational Homo economicus to rest.