The Changing Shape of the British Economy in Recession & Recovery
Headline growth of 0.7% might be a tad down from the 0.8% recorded in Q3 but still represents a solid result. The overall growth rate in 2013 has come in at 1.9% – still below the pre-crisis norm of around 2.5% but much better than the 0.6% the OBR was expecting back in March.
There is no doubt that the economy is growing at a decent clip and growth appears to have picked up in the second half of last year.
But underneath the positive headlines there are reasons for concern. First, as has been widely pointed out, this is not the kind of recovery that we originally wanted. The aim was to rebalance the economy – more exports, higher business investment – and instead we seem to be experiencing what might be termed a ‘typically British’ pattern of recovery with growth linked to a rising housing market, a falling household savings ratio and driven by consumer spending.
The regional picture is much more patchy than the aggregate data suggests and, as of yet, stronger output growth is not filtering through into rising household incomes.
Returning to the good news for a moment, GDP is now ‘just’ 1.3% below its pre-recession peak and on current trends should surpass that peak in the second quarter of 2014 (it might even get there this quarter).
GDP being below where it was six years ago is not something to celebrate, but I have no doubt that the surpassing of that previous level will be widely hailed that as a sign things are on the mend.
Stepping back though, it is worth noting that just looking at GDP levels (whilst superior to looking at just growth rates (e.g. a growth rate of +2.0% is less impressive if it falls one of -4%)) doesn’t give the whole picture.
Recent GDP data and labour market data, when taken together, tell an interesting story. It seems increasingly likely that the severe recession of 2008/09 and the slow recovery since have reshaped the economy.
As I noted last week, whilst the overall number of people in work is now higher than in 2008 the kind of work they are doing is very different.
As the chart below makes clear we have lost 325,000 jobs in manufacturing and gained 425,000 in health and human services. There are 214,000 fewer people working in construction and 115,000 more in real estate activities.
A similar change is clearly visible in the composition of output. As the chart below from the ONS makes clear, the service sector is now above it’s pre-recession peak whilst both industrial production and construction remain below.
Whilst the overall level of GDP might be back where it was in 2008 by this summer, the same will not true of manufacturing output (a component of industrial production) until 2016 on current trends.
In other words whilst the level of GDP may finally be getting back to where it was, the composition has shifted.
I’m increasingly thinking (as I noted over at Left Foot Forward last week) that this big compositional change may explain two of the persistent worries about the UK economy over the last few year – the weakness of productivity growth and the decline in real wages and accompanying living standards squeeze.
James Plunkett of the Resolution Foundation tweeted the below chart yesterday, which appears to give further support to this idea.
It suggests that employment in low paying sectors is well above its previous peak but employment in other sectors remains below.
The UK was always going to face difficult productivity headwinds in the years after 2008. The slow decline of the offshore North Sea sector (a hugely productive sector) coupled with less risk taking from the financial sector (and there’s an interesting and live debate on the extent to which productivity increases in this sector where ‘real‘) knocked away to important components of productivity growth.
But the pattern of the recovery since 2009 suggest that things may be even worse.
Whilst output grew by 0.7% in the three months to December, total hours worked in the economy grew by 1.1% in the three months to November. So either there was a sharp slowdown in hours worked in December (which seems unlikely) or output per hour worked is falling. That implies quite a serious productivity problem.
As I wrote last week:
Back in 2008/09 many expected unemployment to rise by far more than it did. Despite a much more severe recession than in the early 1990s or early 1980, unemployment rose (proportionately) much less than many feared.
Given this one could reasonable expect (and this was certainly the mainstream view amongst economic observers) that any pickup in growth would see weak job growth. The logic was that employers had reacted to a downturn in demand by cutting wages and hours rather than staff and so once the upturn came they could simply increase hours and get more output from their workforce rather than hiring new people.
This hasn’t happened – the recovery over the past year has been employment intense.
So we are left with what economists call the productivity puzzle – output is still 2 per cent below its peak but the number of people in work is higher.
Broadly put there are two distinct ways this could be explained – either we have all generally become less productive at our jobs over the last five years or there was been some sort of change in the composition of the labour market.
It seems increasingly likely to me that the ‘productivity puzzle’ is best explained by a compositional change – more people in low paid, low productivity sectors than before.
That has serious implications for pay levels. Whilst average weekly earnings growth is running at just 0.9%, pay settlements in general seem closer to 2.5%. It may well be that what explains the difference is once again compositional change.
This would mean that people who have remained in the same private sector job over the last few years are starting to see small real terms pay rises – but that people coming into work for the first time, re-entering the labour force or changing job are going into sectors and jobs that, in general, pay less. This drags down the average and fits with the productivity data.
A weak productivity but jobs intense recovery is certainly far preferable to a situation of stagnant output growth and rising unemployment. But it is far from ideal – in the long run lower productivity growth means slower growth in output and lower living standards on the whole.
It might be time for policy makers to start taking the compositional of the jobs market as seriously as they do the headline numbers.