Wednesday, 18 April 2012

The art of looking busy

One of the features of the current depression is that it hasn't been worse. Some of this is down to the support built into the social system since the 1930s, which has helped mitigate the worst impacts of unemployment and low wages, despite the attempts of the government to repeat the macroeconomic blunders of their predecessors and despite their bullying of the unemployed and disabled.

Some of it is down to the broad advance in wealth and living standards over the last 80 years, which has meant fewer visible signs of distress and greater familial resilience, such as adult children moving back in with their parents. Paradoxically, the 30s means test helped break up families, famously in Walter Greenwood's Love on the Dole, while now, when family values are meant to be weaker, unemployment is helping to keep them together.

Some, too, appears to be the result of a general desire to not think too much about the subject. This isn't a case of head-in-the-sand denial, more a wish to hunker down and wait for the shit to pass. Examples of this are the banks' forbearance in respect of commercial loans; building societies forbearance in respect of mortgages; and labour hoarding by firms. The last of these is the least understood, but it may have a bigger impact on the way recovery pans out than the behaviour of financial institutions.

Labour hoarding basically means firms keeping on certain workers, even though demand has dropped, in order to have their skills ready for an eventual pickup. This can be done by reducing hours, so the same amount of work is done by more people, or by keeping wages down (even to the extent of a cut in real terms - i.e. increases below inflation), so that costs can be reduced. On the face of it, this looks like a humane strategy to limit the negative impact of reduced output on workers, which also entails benefits for firms. Not only do they increase staff loyalty, but they avoid costly recruitment fees and learning periods when they wish to expand again. However, there are problems as well.

If headcount remains static while output drops, then productivity must fall as a simple mathematical calculation. Sure enough, there is clear evidence of such a drop, with some estimates putting this as high as 10% below trend level (i.e. the long-run rate at which productivity increases due to technology etc). A comparison of the rate of productivity growth shows this recession to be much worse than those in the 80s and 90s, while estimates of the economy's potential output have taken a major step back, implying a permanent loss of productive capacity rather than just a cyclical downturn - in other words, an L-shaped recession.

I'm sceptical about the latter as estimates of the output gap (i.e. how much more could firms produce with existing labour and plant) are subjective - they're a sentiment, like business confidence. In the services sector particularly, it isn't always easy to predict optimum output, unlike manufacturing where x workers produce y widgets per hour. There is a lot of elasticity. Workers can be diverted to long-cherished projects (rearranging the filing), put on training courses, or engaged in process improvement exercises. Some of this is displacement activity, but some will be an investment in future productivity.

However, if firms have a lot of excess capacity to use up before they need to hire again, this means we could be facing a jobless recovery, i.e. one in which employment growth is weak. This would also mean that youth unemployment would remain stubbornly high, particularly among the unskilled. Recent analysis of the prevalence of jobless recoveries since the 80s points to the way that job polarisation accelerates in these times. This means mid-range jobs that are more susceptible to automation are the ones that don't reappear as vacancies, so that employment growth happens at the top and bottom ends of the job scale, where automation is less applicable.

At an aggregate level, much of productivity growth is down to firms (or offices/plants in large firms) entering and exiting the market. This is partly because new businesses bring new ideas, while failing businesses tend to be unproductive. It is also because new businesses or new plants usually require investment in new technology, which tends to bring higher productivity.

Similarly, a lot of internal productivity growth is down to new hires. Most new hires into the skilled and "knowledge worker" roles that impact on productivity will come from existing jobs. Each recruitment has a cascade effect, creating further vacancies and labour churn. This amplifies productivity growth across the economy as workers are shuffled between employers and new ideas are spread. Churn also impacts on capital investment, notably in areas like IT. New staff usually drive the adoption of new systems, over and above the degree to which the decision to implement new systems leads to the hiring of new staff.

If labour hoarding has been extensive, this will limit the degree to which a recovery will cause a bounce back in labour churn. This in turn may cause productivity growth to remain sluggish. While this may not be significant in specialist sectors like IT, where external factors, such as the adoption of new technologies, help drive churn, it may well lead to slow rates of recruitment in more generalist sectors.

The bottom line is that we may be faced with a recovery characterised by continuing high unemployment and low productivity growth, both as a partial consequence of labour hoarding. This will eventually work its way through, as excess capacity is soaked up and labour churn starts to increase, but that could be many months, or even a couple of years, away. This could stimulate discussion about whether it is time to question how work is shared within society, which would be no bad thing.

1 comment:

  1. A new paper from the Centre for Business Research at Cambridge provides more evidence that labour-hoarding is the chief contributor to lower productivity.

    This also makes the crucial point that the hoarded labour is disproportionately made up of overhead roles, i.e. white collar / middle class jobs, while redundancy mainly affects variable roles that are directly linked to output, such as production line or telesales jobs.

    This helps explain why the predictions in 2009 that the current recession would mainly hit white-collar jobs have (outside the public sector) proved inaccurate. As usual, it's the working class jobs that have suffered most redundancies.