Productivity, real wages & the structural deficit: Why a stimulus now might mean less cuts in the future
This morning the ONS released the latest figures on ‘International Comparisons of Productivity’. They have already provoked a lively debate on twitter between NIESR’s Jonathan Portes and Policy Exchange’s Neil O’Brien. Portes notes that the UK enjoyed a strong productivity performance over the last twenty years whilst O’Brien notes the growth slowed from circa 2002 onwards.
There is an interesting debate on the long term trends in UK productivity – a good place to start is Van Reenen et al’s work on the UK’s growth record, possibly countered by Andrew Haldane’s cautious notes on the UK finance sector’s productivity in the past decade. Van Reenen’s work on the links between productivity and wage growth (summarised by James Plunkett here) are also worth a read, in order to remember this dimension of the debate.
But this post isn’t about long term trends – it’s about productivity in the short run. And in some ways this is one of the most important, although under stated, debates in British macroeconomic policy, one that cuts to the heart of the current disagreements around fiscal policy.
Whatever the long term trends are the UK’s recent productivity performance, post-recession, has been poor. As todays release states:
All G7 countries, including the UK, experienced growth in GDP per worker in 2010, reflecting a bounce-back from the global recession in 2008 and 2009. UK productivity growth was among the lowest of the G7 countries by this measure, while Japan, France and the USA experienced faster productivity growth.
This is well known. The chart below shows the annual growth rate of UK labour productivity over the past twenty years. After a sharp fall during the recession, it has failed to recover and is currently expanding at about half the level it averaged in the decade before 2008 (data from this release).
It’s this collapse in productivity growth which leads to what can be called ‘productivity pessimism’ – a belief that the fundamental drivers of growth have been damaged by the recession, that trend growth is therefore lower and that ultimately the structural deficit is larger (as the structural deficit depends upon an estimate of the trend growth rate of the economy).
Reassessing the size of the output gap and the outlook for growth in productive potential has led us to revise down our estimate of the level of potential output in 2016 by about 3.5 per cent since March.
They also produced a chart demonstrating that productivity growth in the current recovery has been especially weak:
As noted – this matters a great deal. If weak productivity growth assumed to be long lasting then trend growth is lower, the output gap is smaller and the structural deficit is bigger and given the government’s policy framework that means more cuts and tax rises.
Equally if the government assumes (as it seems to) that low productivity grow is constraining the economy then it will pursue ‘supply side policies’ as its primary macro agenda in order to deal with low productivity growth.
But what if low productivity growth isn’t the problem? What if it is merely the symptom?
This is the powerful case by Bill Martin last year, in a report which I’d highly recommend published by the Centre for Business Research at Cambridge University.
Martin notes that both the UK and the USA experienced similar crisis in 2008/09- a banking collapse, a contraction in bank lending, a rapid falling away of demand and a deep recession. But the impact on the labour market was very different.
In the USA unemployed soared, whilst in the UK the adjustment came not from a huge increase in unemployment (although unemployment did rise considerably, it was less than could be expected given the scale of the recession) but rather from a collapse in real wages.
This collapse in real wages maintained employment at a higher level than it otherwise would have done and also supported corporate profits. But – given weak demand, higher employment means (all things being equal) lower output per worker. As Martin explains:
America’s demand deficiency is registered in a high unemployment rate coupled with high productivity. In the UK, greater real wage moderation averted the large increase in unemployment. Demand deficiency was instead registered in the post‐2007 productivity shortfall.
In other words, via the connection of real wages, the UK’s weak productivity growth may be a function of low demand rather a problem affecting the supply side of the economy.
Martin’s analysis ties up with the ONS data. In previous recessions (1990s/1980s) unemployment has risen by a lot more. Where as in this recession the link was less clear, as the ONS have noted:
During the recession, employment fell by less than might be expected given the scale of the fall in output. By the middle of 2009, output was more than 7 per cent below its pre-recession peak, while over the same period, employment fell by around 2 per cent, and total hours worked by less than 4 per cent.
This pattern of ‘labour hoarding’ (and falling real wages) compared to previous recessions may explain the different pattern in productivity growth identified by the OBR.
If weak productivity is a symptom of weak demand and not the cause of the UK’s ills, what does this mean for policy makers? Quite a lot actually.
First – if the problem is a lack of demand, policy should be focussed on boosting demand. In other words, the entire macro strategy of the Government may be worsening the productivity figures.
Second – supply side reforms aren’t likely to be a panacea. If the weakness in productivity is coming from the demand side, then it is unlikely these reforms will have much impact.
Third – if weak productivity is a symptom of low demand then any boost to demand might raise productivity and hence the trend growth rate of the UK economy. Any increase in the trend growth rate should mean a lower structural deficit forecast.
That third point is perhaps the most important. It is often argued that a stimulus now just means ‘more cuts later’. But that might be approaching the problem from the wrong end. If a stimulus now were to boost demand and this increase in demand feed through to higher productivity figures, then the OBR could be in a position to revise up trend output and hence revise down the structural deficit. This is the fundamental paradox here – a stimulus now might mean less (not more) spending cuts are needed in the future.