Over the summer the government replaced its strategy of abstract bluster over Brexit with one of concrete bluster. But there is one area in which the Brexit debate is still being conducted in an air of studied unreality: the fiscal cost.

Before the referendum, the Treasury predicted a £39bn short-term hike in government borrowing in the case of a leave victory, and the collapse of GDP by 7.5% in the case of hard Brexit. But it has now gone quiet.

Before the last budget, in March, the Office for Budget Responsibility (OBR) effectively refused to model the impact on tax and spending until the overall shape of the Brexit deal was clear, claiming there was “no meaningful basis” for any kind of prediction. This allowed the Treasury to act as if there would be no costs.

But in June the Bank of England published a new forecast showing GDP 1.5% lower than predicted by 2020. Meanwhile, the OBR, in a report on long-term fiscal risks, spelled out the danger: if the Brexit process takes just 0.1 percentage points off projected growth over the next 50 years, then without further austerity, debt rises to around 140% of GDP (from 90% next year). It identified three main areas of concern: a slowdown in business investment, a slowdown in immigration and a fall in corporation tax revenues as the City moves to Europe.

It has been highly convenient for the closed and mutually admiring world of the Bank, the Treasury and the OBR to avoid predicting the cost of Brexit, but it cannot go on. Chancellor Philip Hammond is already carrying a £3bn self-inflicted black hole in the current budget, following his U-turn over national insurance contributions and the bung to the DUP. But the real black hole is the conceptual one – writing detailed budgets for an economy whose future nobody can confidently predict.

The headline divorce bill demanded by the EU scarcely matters. Whether it’s the £75bn required or much less, as the OBR points out, it can be absorbed over time. Getting back the money we’re paying in to Europe – estimated by the OBR to be about £8bn once you exclude the rebate and money that the EU spends on farming, regional support and universities – is only positive if the economy does not take a commensurate hit. A Norway-style deal, says the OBR, halves the net benefit to £4bn a year.

But what really matters is whether the UK can grow rapidly – both during a transition period and once the terms of Brexit are settled. But it is here that economics risks turning into ideology.

A report by Economists for Brexit, which predicts strong short-term growth and spectacular future prosperity, has been widely criticised. It predicts the UK economy will expand by 7% after a hard Brexit deal, with four percentage points coming from free trade and abolishing agricultural subsidies, a further 2% from deregulation and the rest from lower welfare payments to migrants. Leave aside whether you would like to live in the kind of Britain the stuffed-shirt Thatcherite dinosaurs advocate – theirs is at least a prediction. But it’s such a huge outlier compared with other forecasts that, if you factor it in at all, it would drag Treasury projections significantly to the positive side.

The latest possible date for the next budget is 13 December. Given Hammond’s track record it is possible that he will attempt, again, to predict the long-term direction of the public finances as if Brexit were not happening. But it would be very unwise. It is one thing to “keep people guessing” as to what your negotiating strategy is – even if the cost is capital flight and a collapse in the supply of nurses and construction workers. But the failure to model Brexit’s impact on public spending and taxation borders would be close to criminal negligence. It leaves – and would be intended to leave – the electorate guessing as to the consequences of the deal until after it is done.

Labour is right to try to vote down the repeal bill this week; it should have no compunction about voting against the final deal either, if it delivers fewer benefits than the status quo.

But in one sense, all politics – including Labour’s – are being formed in a vacuum until the moment either the Bank or the Treasury puts a provisional figure on the growth, tax and trade effects of the deal the government is trying to achieve. It is entirely possible, for example, that in a negative scenario the £49bn of tax Labour wanted to raise to boost health, education and welfare spending might have to be spent covering a post-Brexit collapse in revenues.

And this raises a tantalising possibility. Suppose the Tory government were to fall before the Treasury publishes a single fiscal impact projection. In the snap election that followed, the Treasury would go into purdah. But suppose, the day after the election, a Labour government discovers that the central projection of fiscal losses totally wipes out its social programme. At this point, civil servants would be obliged to tell Jeremy Corbyn: “You can either do Brexit or enact your own manifesto pledges”.

In that situation it would be a no-brainer for Labour to put Brexit on long-term hold and begin negotiations anew. In fact, if you wanted to cancel Brexit, depriving both Labour and the electorate of the true figures until the morning after a snap election would be an excellent way to go about it. That’s why the Treasury’s refusal to predict the cost of Brexit is unsustainable and harms democracy.