Productivity: no puzzle about it
TUC work issued today contests widely-held views that weak growth in productivity is down to failures of skills and/or other defects with the structure of the economy. Our research shows instead that the government’s austerity policies should take most of the blame for the productivity failures of recent years. From a policy perspective, gains in productivity – and hence gains in wages – depend on reversing austerity.
We have a summary piece on the New Statesman blog today. This post gives some background to the debate and then goes through the key steps of the argument. Inevitably, it is quite long (though you could skip the background).
In the aftermath of the financial crisis, the subject of ‘productivity’ has dominated economic debate. In particular, while the government has implemented austerity policies that common sense says should have been harmful to economic performance and individual welfare, economists have instead been preoccupied by productivity outcomes.
Based on output growth divided by employment productivity has been weak for several years (the average since the crisis has been around -1/2 per cent a year compared to 2 per cent a year in the run up to the crisis).
Politicians and policymakers use these results to argue that there is something wrong with the economy in its own right, and these flaws explain wider failures in the economy over recent years, not least the living standards crisis.
There is now a vast literature on explaining potential candidates for this shortfall in productivity. These explanations are primarily concerned with the so-called supply-side of the economy, such as shifting composition from high to low productivity sectors, misplaced capital investment, skills shortages, and dis-functional credit markets. While most commentators recognise demand must play a role, the associated arguments are not well developed. The most honourable exception is Bill Martin’s and Bob Rowthorne’s work of 2012, on which this and previous TUC work has drawn.
Overview of argument
From the perspective of demand, there is no productivity puzzle.
- Low productivity over the period of the coalition government is a result of austerity
- Austerity has led to lower economic growth, which the labour market has been forced to accommodate
- It has done so through price – i.e. major reductions in wages and the quality of work – rather than quantity – with employment numbers holding up
- Low productivity is simply a statistical artefact of this process, a symptom of wider economic conditions rather than a quantity with causal force
The argument has major policy implications. Policymakers currently operate on the basis that productivity outcomes are causal, underpinning both the limpness of the recovery and unprecedented falls in real wages, as well as indicating very limited capacity for future expansion. The government is hence absolved from causing the lion’s share of economic hardship; any stronger recovery depends instead on the workforce and businesses (including banks) getting their act together. In addition, monetary policy is permanently on the verge of tightening.
On the view here, responsibility for the failure of productivity rests squarely with government. Moreover, drawn on the basis of an erroneous interpretation of productivity, judgements about the lack of capacity must be incorrect. This is not to deny supply defects, as in the various contributions to the debate (above), but these must take second place to the necessity of increased aggregate demand to improving outcomes. Reversing austerity will reverse the decline in productivity.
The argument is outlined below, in three key steps (1.growth – 2.labour market = 3. productivity), and through a sequence of charts.
A separate more technical piece looks at the theoretical differences between this approach and that of most other recent contributions. (The only necessary technical background here is that the analysis is based on cash/nominal figures rather than real/deflated ones. First, cash figures are not distorted by the way real terms figures for government spending are based on ‘output’ indicators. Second, productivity outcomes depend fundamentally on the way weakness in economic growth is allocated between wages and employment, which can only be understood in cash terms. Third, when it comes to extending the analysis, goals for policy are based on the public finances which are set in cash terms.)
1. Austerity has meant weak GDP growth
The coalition has reduced the growth of government expenditure. Chart A shows annual increases in government final demand (including wages and salaries, procurement and investment) in cash terms. These have been very subdued relative to increases in previous years: the post-crisis annual increase in spending was £2½ billion a year, compared to the pre-crisis years at around £19½ billion.
A: General government final demand, annual change £ billion
The withdrawal of government spending has meant lower aggregate demand growth and therefore lower GDP growth.
Chart B compares economic growth with government spending growth: GDP growth declines alongside spending growth. GDP growth slowed from an average of 5.0 per cent a year before the crisis (2004-08) to 3.7 per cent after the crisis (2010-13).
B: Nominal GDP and government final demand growth, per cent
The standard way of understanding such changes is through ‘contributions’ to GDP growth. Chart C shows the contributions of the various expenditure components: first, ahead of the crisis, then under the coalition government, and then the differences.
C: Contributions to annual average GDP(E) growth, percentage points
The chart shows very clearly that the whole of the shortfall in growth of 1.3 percentage points is accounted for by the reduced contribution of government (-1.31 of -1.30 percentage points).
The government expected reduced public spending (and ‘hence’ improved public sector finances) to boost confidence in the private sector and lead to a strengthening in activity. But, as the chart shows, this did not happen. Investment spending was only very slightly increased, and consumer demand fell by a reasonable margin. While the OBR has argued that the slowdown in the economy followed from weakening overseas demand in the Eurozone (and associated confidence effects), in both the pre- and post- crisis periods net trade made a fairly negligible contribution.
The reduction in government spending has led directly to the reduction in GDP growth: austerity is therefore the key driver of the poor economic performance in recent years.
2. The labour market accommodates reduced growth through reduced earnings
The reaction of recent poor economic performance on the labour market follows in two basic steps. First, reduced GDP growth is allocated between profits and labour; second, income is allocated within the labour market between earnings and employment.
In the first stage, reduced GDP growth has meant both reduced profit growth and reduced labour income growth. The allocation of reduced income growth between earnings and employment comes from decomposing the total of ‘wages and salaries’ (a national account measure). Wages and salaries is the sum of all employees’ earnings; the growth of wages and salaries corresponds to the sum of earnings growth and employment growth. Again, these are shown before and after the crisis.
D: Decomposition of wages and salaries growth
The chart shows that within the labour market the reduction in wages and salaries growth has ‘simply’ been met (almost) entirely by lower earnings growth rather than lower employment growth.
There is nothing here to be celebrated; workers have had to accommodate themselves to substantially reduced growth in aggregate incomes and gains in employment have been at the expense of lower earnings. This amounts to a major and brutal structural change to the labour market. Work is increasingly of lower quality, concentrated in lower-paid occupations, with higher underemployment: part-time work, self-employment, temporary work, along with a wider lack of security associated with deteriorated contractual conditions and working relations.
While the government claims that recent labour market performance vindicates its policies, in aggregate terms, as with GDP growth, labour income growth – the sum of the parts – has been significantly reduced – it is simply that workers’ incomes have taken a far larger hit than their employment rates.
3. Productivity follows as a residual
Productivity is defined as outputs over inputs, most simply, GDP over employment.
So productivity outcomes here follow from changes in GDP over the pre-and post crisis period against the changes in employment.
Fundamentally, because the adjustment to lower income growth in the labour market has been mainly on the earnings side, the calculation for productivity sets disproportionately higher employment growth against lower economic growth.
The low productivity figures are therefore just a statistical or arithmetic consequence of the manner of adjustment of the economy and labour market, through price (earnings) rather than quantity (employment). This outcome is an effect; it has no causal force.
That’s it, basically; though there’s a lot more detail in the full paper, including some international comparisons that show what is going on in the UK is no freak occurrence.
To re-iterate the conclusion: matters are resolved by the government not acting to reduce aggregate demand, i.e. by increasing spending. Fixing supply problems comes afterwards.
(The same argument permits a fuller and fuller understanding of the impact of austerity on the economy as a whole, not least on the public finances. For example, chart C is indicative of multipliers. This work follows next week.)