Our perceptions of the economy are shaped not just by current reality but also by the past. For example, Ulrike Malmendier and Stefan Nagel have shown (pdf) that people who experienced recessions in their formative years are more risk-averse than others even decades later; workers in the 1950s accepted low wage rises despite a tight labour market because they were cowed by memories of the Great Depression; and I struggled to believe that inflation could stay low in the 90s because my views were shaped by memories of the high inflation of the 1970s.

I suspect a similar motive lies behind the notion that the government must reduce its borrowing soon. Such a view makes sense if you think highish real interest rates are normal - as they were in our formative years. But it's not so sensible when rates are negative.

20 year index-linked gilt yields are now minus 1.6 per cent. This means that if the government borrows £100 now it will have to repay only £72 in real terms. Which of course means that debt can shrink even if the government runs a deficit today. The maths of debt sustainability tells us that we can stabilize the current ratio of government debt to GDP even if the government runs a primary deficit (borrowing excluding interest payments) of around 2.9% of GDP. Its deficit this year will be only 1.1%, according to the OBR.

Richard Murphy is therefore right: there is no need for the government to balance its budget and no need for the forthcoming austerity described by the IFS.

Two counter-arguments to this won't do.

One is that I've assumed gilts yields stay low, which they mightn't do.

True, yields could rise - as the IFS says. But the government has to a large extent locked in lowish rates because it has borrowed long; the average maturity of government debt is 18 years. This means higher yields won't immediately gravely raise debt interest payments. Also, yields are most likely to rise if the global economy proves stronger than expected. But the same stronger activity that raises yields would quite probably also raise tax revenues.

A second argument is that higher government borrowing would raise inflation by strengthening the economy. This, though, is a feature not a bug. We want to get interest rates away from their zero bound, so that conventional monetary policy can act as a cushion when the next downturn comes. Higher inflation is a way to achieve this. There is a strongish case for the macroeconomic policy mix shifting to looser fiscal and tighter monetary policy.

None of this is to say that fiscal policy should never tighten. It should, when real interest rates are at more "normal" levels. When this is the case, there'll be a case for budget surpluses both to stabilize the debt-GDP ratio and to prevent a tightening of monetary policy that pushes rates very high.

Obviously, we are not yet at this stage. Fiscal tightening should thus be delayed.

Such a delay should be used wisely. We should regard it as a chance to better organize the public finances - to ask what sort of tax base we want; what should be the mix of tax rises and spending cuts; and how to find genuine efficiency savings in the public sector - a question which of course requires a knowledge of ground truth which only workers themselves can provide. If such a debate helps to legitimate austerity, it would also help make it more credible and sustainable.

If we had an intelligent politics - which is a big if - a delayed fiscal tightening would be a better tightening.