UK labour productivity - it's mainly a matter of demand

Yesterday the Office for National Statistics published their latest bulletin and data on “Labour Productivity, Q2 2015”.  In it, they note in particular that 

UK Labour Productivity as measured by output per hour grew by 0.9% from the first to the second quarter of 2015 to the highest level ever recorded for this series, albeit some 15% below an extrapolation based on its pre-downturn trend.

Output per hour in services grew strongly in Q2 to a record high, but manufacturing output per hour fell by 0.5% on the quarter, continuing the exceptionally weak trend for this series since the economic downturn.

This “exceptionally weak trend..since the economic downturn” is captured by journalist Ben Chu of The Independent in this chart (posted on twitter @BenChu_), and shows, he says, that since the crash, productivity is some 12% lower than if the pre-2008 trend had continued.

Powerful stuff (though as you see above, the ONS say the gap is  an even bigger 15%).  And it is this stasis in labour productivity improvement over a period of years that has caused so many to fret about the UK “productivity puzzle”, wondering why it has stagnated.

We have long been of the view that – far from being a mystery – the basic answer is that productivity is mainly driven by demand.  And demand has been strongly held back by the government's economic policies which have led to the slowest recovery in recent decades.

But since only supply-side economics are in fashion among the economics mainstream, this issue of demand is largely ignored.  Of course, the UK’s ultra-flexible labour market also plays a role, since it is often easier to substitute labour for capital than in ther countries (see for example my comparison of the UK and France - same GDP and population, vastly different productivity)

While it can be argued that, over time, improved productivity leads to faster-rising GDP, the reverse causation is far more plausible as the dominant factor in any specific shorter period.  Here is a chart made wholly from ONS data showing the annual percentage change on a rolling 4 quarter basis of three elements – output per worker, output per hour worked, and GDP.  The correlation is very strong.  When GDP increases, so does labour productivity.  When GDP falls, and in particular in recessions, labour productivity falls.  

The only significant potentially dissonant point to note is in the period 1990-92, when productivity reached a trough, according to these data, a couple of years before the serious recession.  This merits exploration, but in my view is connected to the fall-out from the wild Nigel Lawson-induced boom.  This apart, the general trends of GDP and productivity are closely aligned. And GDP changes are heavily demand-led, especially in an economy as consumer and property service-focused as the UK’s.

But it is fair to note that there seems also to be another factor at play over the last few years which has held back productivity, though without breaking the broad correlation to GDP – the decline in North Sea oil and gas.  As easily extractable supplies diminish, it becomes more costly and labour-intensive to produces a unit of oil etc., which impacts negatively on productivity – though it also has some negative impact on GDP.

Here next is a chart from ONS which shows the respective contributions of different industrial sectors to changes in productivity.  Productivity is dragged down in particular by “ABDE”, which cover oil and gas extraction as well as other non-manufacturing production, but pushed up by the bulk of “other services” which are the classic consumer demand services, plus areas like real estate that have seen a mini-boom.  

Output in construction and manufacturing has generally risen a little from the trough, but not yet back to pre-crisis levels, and this is also evidenced in the weak but marginally positive productivity figures shown in the chart.

In sum, UK labour productivity mainly changes up and down with the state of the economy as evidenced by concurrent changes in GDP.  This is not to deny the long-term importance for the economy of key supply-side issues such as education and training - but their impact is long-term and indirect.  

The dominant dogma of the economics profession largely ignores demand as a major factor, with its myopic focus on supply-side issues to the exclusion of other factors.  That's why the profession (and its religious disciples in the media) sees only a puzzle, not a solution.