Compared to the recessions of the 1970s, 80s and 90s, the recent crisis in the UK has been characterised by several distinguishing features:
- The fall in output has been more prolonged; the level of GDP did not recover its pre-recession level until the second half of 2013, some five years after the start of the recession.
- However, employment (and total hours worked) have fallen less than the fall in output, and the labour market recovered much more quickly than output has done; employment had returned to its pre-recession level by early 2012.
- Wages on the other hand have fallen by as much as 10% in real terms, and are yet to show any significant sign of increase.
The fact that output has fallen by more than employment (and hours worked) means that productivity has fallen, measured either as output per worker or output per hour worked.
It remains unclear the extent to which falls in productivity are the cause of falling wages or vice versa. Explanations for falling labour productivity include: 1) that the banking crisis impaired the efficient allocation of resources across businesses through reduced bank lending and bank forbearance of bad debtors; 2) industry-specific explanations for productivity decline, such as a tougher regulatory environment, or the need to maintain a minimum operating scale.
Alternatively, it may be the case that labour supply has remained higher in this recession compared to previous recessions due to a combination of changing working practices (such as greater participation of older workers), and welfare reforms which place greater pressure on people to enter the labour market. Higher labour supply, combined with weaker union power, may explain why wages are now much more responsive to output shocks and unemployment than there were in the past. In turn, lower real wages may result in a decline in the capital-labour ratio which helps explain the fall in labour productivity.
So one story that is gaining traction to explain the trends in this recession is that greater labour market flexibility has resulted in the adjustment to falling output coming more through wages, and less through employment, than in previous recessions.
But to what extent does this story hold for Scotland?
- In both Scotland and the UK, output (measured by GVA) fell by 3%, although Scotland seems to have lagged the UK by a year in both the initial fall, and the recovery (Fig 1).
- Employment (the total number of jobs) fell by more in Scotland and has taken longer to return to pre-recession levels (Fig.2). (The large increase in employee jobs in 2014 appears to have been spread throughout the first three quarters of the year, and does not appear to relate purely to a summer spike attributable to the Commonwealth Games). The employment rate has also fallen more in Scotland than in the UK, although it remains below its pre-recession level in both the UK and Scotland.
- The total number of hours worked has fallen even more substantially in Scotland relative to the UK, and in fact the total number of hours worked remains below the pre-recession peak in Scotland (Fig 3).
- Real earnings on the other hand have fallen less in Scotland than the UK, and appear to have stabilised since 2011, whilst continuing to fall in the UK as a whole (Fig 4).
- Putting the output and hours pictures together, productivity – measured as real GVA per hour worked – increased in Scotland in 2008 and 2009 while it was stagnating in the UK as a whole. But since 2009 productivity has stagnated in Scotland too, and in fact fell sharply in 2013 (Fig.5).
Thus, anecdotally at least, there is some evidence that Scotland has experienced a more ‘traditional’ recession than the UK as a whole. In Scotland, the adjustment to falling output has come somewhat more through falling employment and hours worked, and somewhat less through falls in wages.
Structural factors may help explain the slightly different experience in Scotland. Scotland has for example a higher proportion of public sector employment, higher Trade Union density, and lower proportion of migrant workers than the UK as a whole. All these factors may contrive to make wages more ‘sticky downwards’. Scotland may also have a lower proportion of the workforce in micro-businesses, where wage declines have been particularly pronounced. (Although none of these explanations are particularly satisfactory in explaining why wages did fall markedly in Scotland in one year: 2010/11.)
Ultimately, debate about the productivity puzzle is important because of what it means for the policy response. If stagnating productivity is largely a cyclical phenomena, increases in demand are likely to be able to be absorbed by returns to productivity growth. If the causes are more permanent and structural, the implication is that there is less slack in the economy, and attempts to stimulate the economy will only stoke inflation.
One way of looking at how much slack there is in the economy is to consider the Bell/ Blanchflower underemployment index. The underemployment index combines the traditional unemployment rate with a measure of the extra hours of work that those who are already employed would like to work at their current wage. The latest data shows that the underemployment index has followed a very similar path in Scotland and the UK during the crisis, peaking at 10% in late 2011, and since falling to just under 8% (compared to just under 6% before the recession).
One way of interpreting all this analysis is to say that there is a similar level of slack in the Scottish and UK economies, but in Scotland there is relatively more slack in employment and less in wages. This interpretation risks reading too much into the data, but it will be interesting to see how these trends play out.