For me, EU referendum raised a question which nobody has so far asked, namely: if there was a case for a referendum, isn't there a far stronger case for worker democracy?

Many of us were appalled by the atrocious standard of debate in the referendum. Robert Harris called the episode "the most depressing, divisive, duplicitous political event of my lifetime." Perhaps Thatcher and Attlee were right: referendums are a "device of dictators and demagogues."

Which brings me to the question. If you think the referendum was a good idea, you must surely think worker democracy is a far better one. I say so for three reasons:

- In a company, the electorate is much smaller, so one's vote - and that of one's immediate colleagues - is much more likely to matter. This sharpens one's incentives to think clearly and gather evidence. Ignorance and inattention are not so rational in an electorate of 1000 as in an electorate of 50 million.

- In worker democracy, voters have more skin in the game. Whereas in a national election one might hope that the costs of a bad decision are spread across the whole nation, or borne by others, this is less likely to be the case in a company. If it does badly, everyone suffers. Again, this sharpens incentives to get decisions right.

- Worker democracy asks a better question. It asks: "what do you know? How can this company do better?" It's a device for doing what Hayek thought a well-functioning market does - aggregating dispersed and fragmentary knowledge. By contrast, a referendum appeals to prejudice, ignorance and feeling.

On top of all this, worker democracy might have longer-term cultural benefits. For one thing, it would change the nature of management, by forcing bosses to listen to workers. And for another, it might increase the self-reliance and vigour of workers. As Tocqueville wrote:

Democracy does not give the people the most skilful government, but it produces what the ablest governments are frequently unable to create: namely, an all-pervading and restless activity, a superabundant force, and an energy which is inseparable from it and which may, however unfavorable circumstances may be, produce wonders. These are the true advantages of democracy.

And this is not to mention the obvious, and I think overwhelming, case for worker democracy - that it increases both productivity (pdf) and well-being.

All this raises the question. Given that the case for worker democracy is obviously so much stronger than the case for a referendum, how can anyone who favoured having a referendum oppose worker democracy?

The answer isn't that I'm vague about the precise form worker democracy should take. There are countless possible forms, just as there are many different types of political democracy. The appropriate form will differ from firm to firm.

Nor is it because shareholders are the owners of the company and so it is they rather than workers who should have control rights. For one thing, shareholders don't own the firm. And for another, they have in many cases already delegated control to rent-seekers of dubious competence: why shouldn't they therefore do so to more competent agents?

Nor is it good enough to complain that worker democracy might introduce adverse incentives, such as a reduced (pdf) incentive for firms to expand. Actually-existing capitalism has all sorts of agency problems. It is silly to compare worker democracy to some mythical idealized form of capitalism. And if we're talking high theory, the first theorem of welfare economics says that a competitive equilibrium is Pareto efficient. This draws our attention to the possibility that worker democracy plus well-functioning markets might well be more efficient than crony capitalism.

Nor is it acceptable to say there is no demand for worker democracy. This might be due to adaptive preferences, learned helplessness and to management's control of the political agenda, rather than to the fact that worker democracy is inherently undesirable.

My point here is a simple one. Worker democracy should - to say the least - be on the political agenda. That it is not is yet further evidence that politics is systemically dysfunctional.