Why are central banks finding their job so hard to do? A common view is that this is because they are imbeciles. People who assert this insist that central banks need to keep interest rates in line with their historic norms. This is wrong, because historic norms are irrelevant. The questions are why and what this implies for our economies.

A paper by Atif Mian, Ludwig Straub and Amir Sufi at the Jackson Hole monetary conference on 27 August illuminates this issue. It reaches a conclusion, already suggested in their earlier work: the principal explanation for the decline in real interest rates has been high and rising inequality and not demographic factors, such as the savings behaviour of the “baby-boom” generation over their lifetimes, as some have argued.

The analysis starts with estimates of the real “natural rate” of interest, a concept that goes back to the Swedish economist Knut Wicksell. The natural rate, he explained, balances demand and supply in the economy, which shows itself in stable prices. The modern doctrine of inflation targeting has descended from this idea. Crucially, however, estimates of this rate for the US show a fall from about 4 per cent four decades ago to around zero now.

This decline is matched in other high-income countries, as one would expect: in an open world economy, equilibrium real interest rates should converge. As the paper also notes, the decline “raises concerns about secular stagnation, threatens asset price bubbles, and complicates monetary policy”. Indeed, it is a big part of the reason why central banks have had to make huge asset purchases in crisis situations, such as now.

Their main point is that savings rates vary far more by income within age cohorts than they do across age cohorts. The differences are also huge: in the US, the top 10 per cent of households by income have a savings rate between 10 and 20 percentage points higher than the bottom 90 per cent. Given this divergence, the shift in the distribution of income towards the top inevitably raised the overall propensity to save. As an explanation of rising propensities to save and the falling real interest rate, the shift of the baby-boom generation into middle age does not work, because rising savings have been continuous while the impact of the demographic shift on savings behaviour has not. (See charts.)

At the aggregate level, savings must match investment. So what happens when the rich get richer and so try to save more? Interest rates must fall. It turns out that the impact of this on business investment is quite feeble. Indeed, the propensity to invest has been chronically weak, partly for demographic reasons. So the offsets have had to come either from persistent fiscal deficits or from higher spending by the bottom 90 per cent. Both are fuelled by debt, while the latter is also powered by asset price bubbles, especially in house prices. As central banks pursue the natural rate downwards, they drive both of these processes. But, as debt ratios rise, natural rates fall still further, as the highly indebted become ever less creditworthy.

An objection to this argument is that it is just about one country, however important. But the tendency towards more income inequality is shared by almost all large economies, including notably China. Indeed, the excess savings of the rest of the world have also shown up in persistent US current account deficits. The need to offset the latter has made the task of the Federal Reserve yet more difficult.

The financial crisis of 2007-12 should be seen as an outcome of these processes, resolved at the time by rescuing the financial system, tightening regulation and doubling down on low rates across the yield curve. The Covid crisis was a bolt from the blue but the response was more of the same, but on an even bigger scale. This time, moreover, the huge increases in central bank reserves actually increased broader monetary aggregates. It is no great surprise, therefore, that the combination of supply side disruptions with today’s strong demand are generating “surprise” inflation.

So how might the story evolve? There is no powerful reason to expect income inequality, the fundamental driver of today’s excess savings, to reverse, although it might stabilise. There is an excellent reason for a huge investment boom, notably the climate transition. But that will not occur without consistent, determined, intelligent and globally aware policymaking, none of which we can expect, though we may hope. So, in the medium to long term, secular stagnation is likely to return, unless income inequality falls.

The short term is harder to read, but if it goes wrong, is disturbing, perhaps even for the medium term. In his speech at Jackson Hole, Jay Powell, chair of the Federal Reserve, insisted that all is under control. But he would say that. The surge in inflation has in fact surprised almost everybody. The worry must be that the price shocks persist and then get baked into expectations, which will then only be reversed by a period of significantly higher short-term rates. That would cause stagflation, which would create painful dilemmas for central banks and surely cause devastating problems for weaker borrowers, notably but not exclusively heavily indebted emerging economies.

The exceptional policies of 2020 can no longer be justified. Given today’s super-low short-term interest rates and supportive fiscal policies, it is hard to see why large asset purchases should continue, either. We have more than enough money today and bond yields ought to rise a little. When the facts change, central banks should change their minds. That time is now.

martin.wolf@ft.com