Duncan worries that economists might be 'misdiagnosing a serious demand problem as a productivity problem.' It could be, he says - endorsing Bill Martin's view (pdf) - that productivity growth has been weak recently not because of a supply-side problem but because weak aggregate demand has caused firms to hoard labour.

This is not merely an abstruse technical argument. It matters politically. The view that we have a supply-side problem has two implications:

- it implies that the output gap is small, which in turn implies that a large part of government borrowing is structural rather than cyclical. The case for fiscal tightening is thus stronger.

- if our problem is on the supply-side, then the policy remedies are less likely to involve fiscal expansion and more likely to entail supply-side reforms which - given the dominant neoliberal ideology - mean incentivizing bosses and bashing workers.

But are Duncan's worries correct? Two things make me sympathize with him. One is that I'm an old git who's been here before. I remember in the early 80s worrying that the then-mass unemployment would lower future growth through hysteresis effects. And I remember the better times of the late 80s and late 90s leading people to revise up their estimates of trend growth. History tells me that estimates of potential growth are pro-cyclical.

I'd add another reason why this might be so. It could be that the banking crisis, allied to the cyclical element of firms' reluctance to invest, has slowed down the entry (and exit) of new establishments which is a major cause (pdf) of productivity growth.

On the other hand, however, three things make me doubt this:

1. Productivity growth was slowing down before the recession. In the five years to December 2007 it grew by less than 2.3% a year. That's below the post-1990 average, despite what should have been a boost from a cyclical upswing.

2. Inflation has consistently been quite high since the recession began. This is consistent with the possibility that the output gap is indeed small - though as Simon says, there might be other reasons for this.

3. Even insofar as low productivity is a result of the recession, it doesn't necessarily follow that fiscal stimulus alone will raise it. If low productivity growth is due to the failure of the banking system to promote the start-up and expansion of new and (potentially) high-productivity business units, then a fiscal expansion that is unaccompanied by banking reform might not greatly raise productivity.

Rather than take a Keynesian or neoliberal supply-side view here, can I suggest an alternative? It's that, insofar as the slowdown in productivity growth is not cyclical, it reflects not the problems traditionally identified by neoliberals, but others, such as:

- the dearth of monetizable investment opportunities has lead to slower capital formation which inevitably means slower productivity growth.

- there are limits to the extent to which top-down managerialist organizational structures can identify productivity improvements and motivate workers, and we are bumping up against these limits.

- the UK banking system has always been poor at facilitating productivity-enhancing activity, and it is especially so now.

Insofar as these factors lie behind the productivity slowdown, the solution is neither Keynesianism nor orthodox supply-side reforms, but something else.