Whom should we trust: elites or the people? I'm in two minds here.

On the one hand, on the issues of immigration and Brexit, I'm with the elites.

But in other respects, I'm on the side of the people. I've repeatedly argued for worker democracy against managerialism, and have consistently pointed out the shortcomings of professional fund managers. And I've pointed out that popular opinion - as measured by ratios (pdf) of consumer spending to asset prices - can do a better job of forecasting future economic conditions than professional forecasters can manage.

So, am I just confused?

No. There's a logic to my position which is lacking in extreme pro- or anti-elitism. To see it, remember that crowds are wise only under specific conditions. These are:

- Beliefs are uncorrelated, so there's no herding or information cascades.

- There's a diversity of belief, with people drawing on their personal and perhaps tacit knowledge of local circumstances.

- People have incentives to get the decision right.

I side with the people when these conditions are in place. This is true of the ability of consumption-wealth ratios to predict future economic conditions. This works because each individual decides his spending according to personal circumstances: is my job safe? Will I get a decent bonus? Do I have good prospects? And so on. And everyone has an incentive to get the decision right: spending too little means unnecessarily depriving yourself whilst spending too much means uncomfortable retrenchment later. Granted, it's possible for spending decisions to be influenced by peer pressure. But I suspect this mechanism is often not powerful enough to offset the strength of the other conditions.

I suspect too that these conditions would often facilitate worker democracy. Each worker has local specific information about how the firm might increase efficiency, and has skin the game (job-specific human capital) that incentivizes him to express a good opinion.

However, in the case of Brexit, these conditions are absent. Voters' beliefs are correlated, by virtue of being unduly influenced by the shit media (including the BBC) as well as by cognitive biases. There's little local, fragmentary knowledge to be aggregated: not many voters were swayed by their personal experiences of dealing with the EU. And they didn't have skin in the game: many older voters believed, sadly correctly, that the costs of Brexit would be borne by others.

Herein lies the paradox. Although there exist good criteria for adjudicating when we should trust the people and when the elites, I seem to be the only one who applies these criteria. Instead, popular opinion distrusts elites when they should be believed - for example on Brexit - and yet believes them when they should be distrusted: there's insufficient demand for worker democracy, and far too much trust still placed in fund managers.

What we're seeing, therefore, in today's populism is the exact opposite of what intelligent, evidence-based class war should be.