Is Chancellor right to want surplus in normal times?
Although George Osborne committed the Tory party to a new fiscal rule before the general election - that the government would generate surpluses in all so-called "normal times" - it matters that he will tonight be confirming this rule is to be enshrined in law.
Because this represents a significant shift to small "c" conservatism in the stewardship of the public finances, compared with the consensus of the last 70 years.
Typically governments have either had no such statutory rules at all, or have been subject to looser constraints - such as the last Labour government's commitment to the "golden rule" of borrowing only to invest over the course of the economic cycle, which in practice wasn't a huge constraint on its public-spending ambitions.
So the chancellor's surplus promise represents a significant shift in economic and political judgement, and not just that the current national debt - equivalent to about 80% of GDP or national income - is too high and should be cut.
What the chancellor is implicitly arguing at the Mansion House tonight is that the public sector cannot generate the growth necessary to get the national debt down faster - that the optimal path is for the public sector to shrink, creating space for the private sector to take up the slack.
Now for the avoidance of doubt, quite a lot of economists - most but not all of them leftish - would say this is bad economics.
They might accept in formal terms, that national debt as a share of GDP would fall a bit faster. But they would also claim that we would be poorer as a nation, because what they would see as debt-reduction fetishism would mean that growth and prosperity would be lower than would otherwise have been the case.
So the new fiscal law will put the Labour Party, currently without a permanent leader or an ideological direction, on the spot.
It will be a test of the leadership candidates' commitment to maintain the size of the state. And it should occasion an unusually lively and important debate among wonks and policymakers.
In case you wondered, running a surplus in most years does not fix the public finances overnight, on the Treasury's calculations.
It estimated at the time of the last budget that if the government ran a 0.3% surplus on average it would take at least 10 years for the ratio of debt to fall to 60% of GDP - which is the maximum of the EU's Maastricht treaty, and therefore one benchmark of prudence.
And for debt to fall to roughly where it was before the 2007-8 banking debacle and recession, or just over 40% of GDP, would take two full decades.
Mending the roof, in George Osborne's favourite phrase, is not apparently a job apparently capable of being done at the speed of an emigre Polish builder.
That said, the Treasury's projection is that the national debt would not drop to that Maastricht level of 60% or so till 2036, if the government continued the previous approach of borrowing to invest, of running a surplus only on the current budget rather than on the overall budget.
So Osborne makes two other points.
First that we can't be certain there isn't another 2008-style economic calamity lurking just round the hairpin bend. And if we slump when our national debt is close to the current 80% of GDP, we'll then see government debt increasing to the dangerous growth-suppressing levels of an Italy (or worse).
Second, that the example of prudent Canada and Sweden is that public services are less vulnerable to savage cuts in the long term if the public finances are kept in healthier shape, because it means there should be more scope to borrow very temporarily, to cushion them in a downturn.
And there indeed is the heart of the matter - which is what kind of flexibility a chancellor would have under his approach to take evasive action when there are shocks, and also what on earth is meant by "normal" economic times?
Today it is impossible to properly assess the new surplus law, simply because we don't know the mechanics of it - we don't know how tied the chancellor's hands will be in practice.
There are a number of crucial uncertainties.
First, is "normal" growth in GDP about 2.6% or so - the average rate of growth since 1945? Or in the wake of the productivity-smashing disaster of 2008, is underlying growth now nearer 2%?
And would the de facto target for growth be a point, say 2%, or a range - a bit like the inflation target - of 2% plus or minus one percentage point?
Also would the direction of growth matter? In other words, would there be an expectation that surpluses would typically be generated when growth is rising, but that it is OK to borrow when growth is falling?
Now before the election the chancellor said that he would expect the Office for Budget Responsibility to be the arbiter of "normal" times.
But what does this mean?
Presumably the OBR will determine the UK's "normal" rate of growth, every few years or so. But will it then decide at any particular moment whether current growth is sub-normal or supra-normal?
And how often would it judge whether growth is normal and expected to remain so? Just twice a year, at the time of the autumn statement and in the Budget? Or would it be expected to make snap judgements as and when there is a big shock to the economy.
What matters probably most of all is whether the OBR would make the judgement about whether times are normal in retrospect, over the course of an economic cycle, in determining whether the Treasury has flunked or passed its surplus rule - or would it do this in a forward-looking way that could be prescriptive of the chancellor's actions?
So although the idea of generating a surplus in normal times may seem to have the virtue of simplicity, in practice it may be fiendishly complicated.