The reason the pay debate goes nowhere is that it is predicated on the most stubborn and damaging myth in business: that shareholders own companies.

Once and for all: they don’t. According to two law professors writing in that revolutionary organ, Harvard Business Review, ‘the law provides a surprisingly clear answer: Shareholders do not own the corporation, which is an autonomous legal person.’ Directors’ fiduciary duty is to the company, not shareholders. Shareholders own shares, which give them rights to residual cash flows and to vote on resolutions and board elections at the AGM. They have no ownership of the company’s assets, which are owned by the company. They don’t even have an unqualified right to dividends–the most valuable company in the world, Apple, rich enough to buy the eurozone, has never paid a dividend in its life. Shareholders, says Charles Handy, our most eminent business philosopher, no more own companies than a punter on the 2.30 at Tadcaster owns the nag he is betting on.

This is not just a semantic difference. Although you’d never guess it from the acres of writing on the subject, high pay isn’t an aberration of the system but its predictable outcome–the logical creation of governance arrangements that assume that shareholders are the boss and it is the manager’s job to do their bidding.

Where did the myth come from? For once it is possible to pinpoint the source with some accuracy. Flashback to the end of the 1970s, when a growing feeling that shareholders were being short-changed by corporate managers who had grown fat and lazy in the long post-war boom was crystallised by Michael Jensen and William Meckling in a paper that despite its less than pulse-quickening title, ‘Theory of the Firm: Managerial Behaviour, Agency Costs and Ownership Structure’, remains the most quoted ever in the economic literature.

The authors cast managerial underperformance as a ‘principal-agent problem’. In their construct, shareholders are the firm’s ‘principals’ who hire managers to run the company on their behalf. The ‘agency problem’ arises because if they can get away with it managers will (like everyone else) put their own interests first. In other words, the incentives are misaligned. Ergo, the argument runs, the way to get managers to do their job is to realign the incentives by giving them significant amounts of stock-based compensation, turning executives into shareholders too.

There is one glaring snag in the theory. In law, directors and managers aren’t employed by shareholders at all but by the company as ‘autonomous legal person’. But that doesn’t square with the new (and wholly ideological) assertion that the company’s sole purpose is to maximise shareholder value. That can only be the case if shareholders actually own the company. To sidestep this inconvenient fact, the authors simply dismiss the company’s autonomous status as ‘legal fiction’–a ‘simple falsehood’, points out Gordon Pearson in his careful study, The Road to Cooperation, on which is based the entire edifice of governance that has stood ever since.

Ironically, Jensen and Meckling’s pro-shareholder remedies were eagerly seized on managers who correctly spotted in them a bonanza-in-the-making that would make their previous pickings look like small change. The theory expected them to be greedy; they complied in full, demanding ever greater incentives for their alignment in a perfect example of the self-fulfilling prophecy.

The full accounting for the era of ‘shareholder primacy’ has yet to be done. But we do know that, as Roger Martin has shown, shareholders have paradoxically done less well since the 1980s than in the previous 30 years when managers were supposedly feathering their own nests. A little reflection shows why. Doing exactly what the incentives told them to, a new breed of imperial CEOs used every means in their considerable power to jack up the share price (and thus their own salaries) in the short term. They bought back their own shares, did deals and financially reengineered their companies. They also cut spending on research (so innovation stalled), slashed the workforce and dumped pensions. The sufferer was the company, the true principal, looted by an unholy alliance of insider-managers, both corporate and financial, who hijacked the spoils. The effect of the Jensen/Meckling model has been the exact reverse of what was intended.

Now we can see why halting the CEO pay up escalator is impossible without disposing of the ownership myth. Present-day governance based on it is a recipe for increasing CEO pay whatever their performance. Calls to cut or even abolish bonuses are irrelevant, since boards will simply transfer the ‘incentive’ to another kind of pay. It’s the legally erroneous and counterproductive incentive to maximise returns to shareholders that’s the issue, making CEO pay rises not just, in J K Galbraith’s words, ‘a warm personal gesture from the individual to himself’ but one that is justified by the official version of how the company functions.

Once the ownership issue is resolved, everything else falls into place. When it is recognised–and the 2006 Companies Act could hardly be clearer–that directors’ duty is not to shareholders but to ‘promote the success of the company for the benefit of its members as a whole, and in so doing have regard for… the long term’, then long-neglected issues of internal fairness and morality come bursting in from the cold, bringing an icy blast of reality with them. For example, Peter Drucker told Forbes magazine back in 1997 that top pay of more than 20 times the average hourly wage could only damage company morale. Few top executives, he added, ‘can even imagine the hatred, contempt, and even fury that has been created – not primarily among blue-collar workers who never had an exalted opinion of the “bosses” – but among their middle management and professional people.’

Since then that multiple has soared to 145 times, and much more in the US. That is the reason for the ‘hatred, contempt and fury’ that has spilled into the open these last few days and so surprised RBS’s Hampton and Hester. One excellent reason for putting employee representatives on remuneration committees is to make sure they won’t forget it again.

In a recent article the FT’s Martin Wolf observed that the chief failing of that ‘brilliant social invention’, the limited liability company, was that ‘it is not effectively owned. This makes it vulnerable to looting’. Exactly. He confessed that he was unable to come up with a remedy. But the answer is staring us in the face. Ownership is a red herring. Instead of trying to make it work, all we have to do is amend the governance codes to reflect the reality incorporated in the 2006 Companies Act rather than the myth generated by a 1976 article in the Journal of Financial Economics–the company is the principal, not shareholders.