Productivity: It's a gas

Last November I experienced a leak in my gas supply which required the replacement of a pipe. All seemed to be well until last week when I again smelled gas. Having tried to call the original engineer to rectify the problem – to no avail – I contacted a different local engineer. He discovered that the person responsible for installing the replacement pipe had allowed soldering flux to drip onto it, despite the fact that the instructions for fitting such pipes advised covering it to avoid any contact with flux, with the result that it had started to corrode and required replacing. Away went the gas engineer to get a new pipe, only for him to find that the one he obtained did not fit which necessitated a second trip to the wholesale supplier.

In the end the job was successfully completed, but what should have been a simple one hour job last November turned out to require four hours of work spread over 8 months due to the shoddy initial job and the failure of the second engineer to measure the pipe before purchasing a replacement. This got me thinking about productivity and the reasons why it is so low.

Compared with the rest of the G7, UK productivity lags behind. On data to 2016, output per head in real terms across the other G7 countries was almost 6% higher than in the UK with German output per head 4% above UK levels and the US 7.4% higher. A study conducted by NIESR for the Department for Business, Innovation and Skills in 2015 concluded that although the UK performs well in an international context on the basis of the contribution of higher skills (i.e. university education) to productivity, “the UK’s intermediate (practical, technical and occupational) skills are of more concern.” This would accord with my own anecdotal experience in which an apparent lack of training resulted in poor basic skills with the result that fairly simple jobs take longer to complete than necessary (not to mention costing more money).

But it is not just a British problem. Policymakers have been exercised in recent years by the weakness of productivity growth, which has been one of the marked features of the recovery from the global financial crisis of 2008-09. Only 6 out of 36 OECD countries recorded faster productivity growth over the period 2010 to 2018 compared with the years 2002 to 2007. Across the OECD as a whole, labour productivity has averaged growth of 1% per year since 2010 versus 1.7% between 2002 and 2007. In the euro zone, where productivity growth was not especially rapid before the GFC, it has slowed from an average of 1.1% per year to 0.8% whilst in the US the slowdown has been much more pronounced (from 2.1% to 0.9%). The slowdown in the UK has been almost as severe, leaving the rate of productivity growth at a paltry post-GFC average of 0.6% per annum versus 1.7% in the five years prior to the recession.

Aside from impact on wasted time and higher outlays, why do we care? In the first instance, productivity is the key driver of living standards. In an economy where labour is paid roughly according to the amount of value added it generates, there is a strong link between (real) wage growth and output per worker. Thus, if productivity growth slows, so too will real wage growth and over the period 2008 to 2014 real wages in the UK fell by 9%. The “puzzle” in all of this is that the sustained general pattern of productivity stagnation contrasts with the pattern following previous economic downturns, when productivity initially fell but subsequently recovered towards the previous trend growth rate. Something has clearly changed but we cannot be sure what it is.

Chart 1, 2 and 3 show UK GDP, productivity (output per worker) and employment trends (respectively) in the current cycle compared to  the three major recessions since the 1970s (click to enlarge). The current GDP upswing broadly matches that of the 1970s and early-1980s but lags the 1980s and 1990s. But employment growth has far outstripped previous recovery phases whilst productivity has lagged. To the extent that output is a function of labour input, capital investment and technological progress, it appears that British companies have increased output by expanding labour input rather than capital investment or technological improvements. If these workers are lacking in the appropriate training, it is likely they will take time to get used to the systems and processes required to enable them to do their jobs efficiently with consequent adverse effects on productivity.

It also appears to be the case that there is a widening gap between the performance of firms in the upper quartile of the productivity distribution and those lower down. One reason for this is perhaps that the best educated workers tend to gravitate towards the best paying firms who then improve the company’s productivity performance and are then rewarded for their efforts. In other words, there is a self-reinforcing spiral which rewards those who already have the requisite skills. What is thus required is a more efficient transfer of skills throughout the economy. This is a problem which has long been recognised and the government’s apprenticeship scheme is a way to help give decent vocational training to younger workers (though changes to the Apprenticeship Levy in 2017 may be hampering the operation of this market).

However, there is no quick fix for poor productivity. In any case, measuring it is particularly difficult in an economy in which services account for an increasing share of output. Last December the OECD had a closer look at the way in which it measures productivity and concluded that once we account for differences in the way we measure labour input, particularly with regard to the way in which holidays are treated in the data, the UK’s international productivity does not lag quite as much we had previously thought. Ironically, if my original gas engineer had been on holiday when I called him last year, maybe I would have been spared the result of a lack of training. There again, I would have missed out on a learning experience.

The great productivity debate remains unresolved

Over the course of the last week I have listened to presentations from three of the UK’s most influential forecasting institutions (the Institute for Fiscal Studies, National Institute of Economic and Social Research and BoE) in which the common factor was the weakness – indeed lack – of productivity growth. The ONS measures productivity using three main metrics – output per worker, per job and per hour. In the nine years since the great recession, these have grown by 0.8%, 1.0% and -0.2% respectively. For the avoidance of any doubt, these reflect the absolute changes across a nine year span, not the annual average – in other words, output per hour has declined in absolute terms by 0.2% since Q4 2007 (do you feel as though you are working less?).

To put this into context, annual growth over the period 1980 to 2008 averaged 2% for output per worker; 1.3% per job and 1.4% per hour which means that output per worker is currently almost 13% below its pre-crisis trend (chart). You really have to go back a long way to see a sustained period of sluggish productivity as poor as this in the UK during peacetime – probably back to pre-industrial revolution times, if the BoE’s excellent long-run historical dataset is any guide.

On the assumption that real wages grow in line with productivity, this means that real wages today are a similar amount below where they would otherwise be. But what makes matters worse is that, miserable though productivity growth has been, UK real wages have lagged even further behind. Average weekly earnings deflated by the CPI are currently 4.6% below their end-2007 levels. In other words, productivity growth has been flat but real wages have fallen. This goes to highlight the claim made by Paul Krugman that productivity may not be everything, but in the long run it is almost everything. It is undoubtedly one of the reasons why we all feel so much worse off than we did prior to the recession.

None of this is exactly news, as it has been a mounting cause for concern in recent years. Indeed, I recall looking at the problem as long ago as 2013. But the fact that productivity has remained stubbornly weak, and not recovered in line with expectations (and indeed, historical experience) clearly points to something wrong. It might be partly the result of measurement error in an increasingly digital economy; labour hoarding in a service driven economy; a lack of investment or simply an ageing demographic profile. It is, of course, not just a UK problem: It is a problem across all industrialised economies although the UK seems to have been hit relatively hard. Nonetheless it is an issue which governments have tended to downplay up to now.

Each of the three UK institutions noted above sees different aspects of the problem, reflecting their particular speciality. For the NIESR, it will contribute to weak growth and thus an ongoing squeeze on living standards. As far as the IFS is concerned, whose focus is fiscal policy, the resultant weakness of growth will put pressure on the economy’s ability to generate sufficient revenue to eliminate the budget deficit. Indeed, the Office for Budget Responsibility last month noted that it is likely to “significantly” reduce its “assumption for potential productivity growth over the next five years“ in its November forecast which accompanies the Autumn budget statement. Meanwhile, the BoE is concerned that weak productivity growth will hold back potential output growth with the result that the economy is more likely to run into capacity constraints which will generate inflation.

So far, we seem to have escaped the worst aspects of sluggish productivity growth. Inflation has only picked up because of a sterling-induced crash, not because of any underlying rise in domestic cost pressure. Public finances have improved significantly, with public borrowing declining by two-thirds between 2010 and 2017. Arguably the squeeze on living standards has been the worst aspect, and may well have contributed to the Brexit vote last year. But the sunny uplands to which politicians promise we will soon return will remain unattainable without a recovery in productivity. And the longer productivity remains weak, the more of a struggle economic recovery will be. Like a Facebook relationship status, the productivity puzzle is complicated but it is one we need to get a handle on – and fast.