What explains the change? It’s partly a tribute to Britain’s success: our official unemployment rate is the lowest in the European Union, hovering at 1975 levels below a million. We no longer think of joblessness as a problem. But there is a deeper explanation, too.

Industry mattered to politics when politics mattered to industry. Two decades ago, whole sections of the economy were under the direct control of the state. Now, in the era of privatisation, ministers are able to shake a fist or shout the odd plea from the sidelines but rarely to make the decisive difference. They can beg BMW or Motorola not to pull out of Britain – but if the boardroom has made up its mind, there’s little even a phone call from the PM can do.

So voters no longer look to government to make the industrial weather. Two decades of Thatcherite economics have persuaded us that the market is king: governments are powerless to resist. On this logic, politicians have a choice between doing nothing or making things worse. Their role is to stand aside and let the market sort it out. We are all laissez-faire liberals now.

And so the very phrase “industrial policy” – such a staple of 1970s political talk – has disappeared. And yet it’s worth examining Labour’s stance on industry, for inside it lies a glaring, increasingly risky contradiction – and Black Monday illustrates it perfectly.

The government believes in blending the lean efficiency of the private sector with the social goals once exclusively associated with public ownership. The result is a new industrial landscape littered with strange, hybrid creatures – part private, part public, they look and behave like neither. They are the “third way” made flesh.

Railtrack was one. Inherited from the Major government, this company walked like a private business, talked like a private business – but never quite escaped its genetic origins in the public sector. So it had shareholders and sought profit, but as soon as things went wrong it held out the begging bowl for handouts from the government. Of course it got the money: how could any country let its rail system go under? So Railtrack had all the fun and perks of life as a private company – safe in the knowledge that, whenever the chill wind of the market got too nippy, the nanny state would be there with a blanket.

That’s why so many voters, commuters especially, feel resentful about forking out an extra £500m in compensation to Railtrack’s shareholders. They know that reason is probably on the shareholders’ side: they owned assets which the government could not simply grab from them. But the admittedly emotional response of many is to ask: if your company was worth so much, how come you kept coming to us for more cash? More viscerally, why should we bail you out, just because your shares went down; how much did you give to us, the taxpayers, when they soared up? And isn’t that the whole point of shares: you do well if they go up, but you expect to take a bath if they go down? Railtrack shareholders placed a one-way bet: win if you win, but don’t lose if you lose.

And this is the core problem: we were asking a private company to take on an essentially public task. Railtrack was designed to follow the profit instinct of private enterprise when its real job was to provide a public service. It was meant to be governed by the iron disciplines of market forces, but it always knew its risk was more hypothetical than real: if trouble struck, the government would step in – as Stephen Byers duly did this week.

There is a direct lesson here for the public-private partnership plan still dogmatically pursued by Labour for the London Underground. Once again, the government will pretend that the private infrastructure companies are taking on the risk that things might go wrong. They will certainly be handsomely rewarded for it. But if things do go off the rails, we all know who will really pay the price: the government cannot let the tube collapse, so it will step in – with our money.

Consignia is a different strain of mutant company. It too was meant to behave like a private outfit, even though it remains government owned. And it, too, suffers for being neither fish nor fowl. It faces competition for key services, like a private company, but it cannot do what any private business would do if strapped for cash: it cannot raise the price on its core product. The price of a first-class postage stamp has gone up just once in six years, even though it costs a penny more to deliver a letter than it costs us to post it. The regulator has capped the price, on the reasonable logic that monopolies can’t just up their charges whenever they like: after all, the customer has nowhere else to go.

So the Post Office is sort of private, sort of public: exposed to competition, yet obliged to perform public duties (like delivering letters to remote rural locations) that cost them badly. Its rivals are full-blooded private businesses, able to cherrypick the profitable bits, unhindered by costly obligations. The Post Office is neither one thing nor the other – and soon 40,000 workers will pay the price.

The government needs to have a rethink. It should follow the logic of Gordon Brown’s speech last week on the NHS, and declare that some tasks are public by their very nature. Health is one, said the chancellor. Why not add railways, which will always require a public subsidy, and a collective, social need like delivering mail?

“The plain fact is, there are certain natural monopolies, best run by the state,” says director of the Industrial Society, Will Hutton. That does not mean, he adds, that they have to be run like the “organisationally dysfunctional” nationalised industries of the 1970s. Network Rail, the successor company to Railtrack, could be a step in the right direction. Its directors will be rewarded not for boosting share price, as with Railtrack, but by their performance on the “public” aspects of the service: safety, reliability, punctuality.

That may be a new way of doing things. But only if the government ends this unhappy experiment in asking private companies to do the public’s work. That experiment has failed.