Productivity isn't everything

“Productivity isn’t everything, but in the long-run it’s almost everything.” Everyone knows, and agrees with, this sentiment of Paul Krugman’s.
Which is a shame, because it isn’t true. This new paper shows that for several developed economies - including the UK - there is no common long-run trend between GDP and productivity growth.
Even over long periods, GDP growth can be significantly affected by changes in labour utilization, not just by productivity.
Take, for example, France and the UK. The authors show that, between 1980 and 2005, the UK grew faster than France. But France had the better growth in hourly productivity. Instead, the UK’s superior growth came partly because hours per worker did not fall so far and partly because the employment rate rose.
Here, I can hear Tim Worstall: “jobs are a cost, not a benefit.” This is only half right. It’s right in the sense that if people choose not to work, or to work less, then there is nothing necessarily wrong with a slowly growing economy.
It is wrong, though, if people would like to work more, but cannot do so. In such circumstances, UK-style growth - through employment rather than a productivity surge - can be welfare-enhancing; unemployment, remember, is a massive source of unhappiness.
This has important policy implications. It suggests that measures to improve labour utilization, ie to achieve fuller employment, can significantly improve GDP growth not merely over cycles, but over the longer term - that is, insofar as anything can.
This is not to say there is a trade-off between productivity growth and employment growth. Quite the opposite. One way of raising labour demand is to increase productivity.
What it does mean is that, when there are six million people who are unemployed or under-employed, it is a little odd to think that long-run growth requires only productivity growth rather than increased employment.
Productivity is not everything - which, given the UK’s recent performance is just as well.

Comments

[alastair harris]

"It is wrong, though, if people would like to work more, but cannot do so." Wrong in what sense? And in what sense do you mean "cannot do so"? There are certainly job vacancies available, although it may be that the jobs or the pay rates on offer are not attractive. Are you suggesting a universal right to work? This is wrong on so many levels!!!

[Luis Enrique]

this is a nit pick ... accepting what you say about importance of involuntary (or voluntary because what's on offer is crap) non-utilization of labour, which you are right to complain is neglected ... that paper doesn't really show productivity "isn't everything" in the sense I think Krugman means, because it may be high productivity that allows people to choose low labour utilization, so even if productivity doesn't predict GDP, it may predict what matters (a higher sum of leisure and output).

[Andrew]

And lo, a productivity post was commented on by productivity-enhancing (?) spambots...

[ortega]

Oops!

http://online.wsj.com/article/SB10001424052748703858404576214582423794562.html

[rjw]

"This is not to say there is a trade-off between productivity growth and employment growth. Quite the opposite. One way of raising labour demand is to increase productivity".

Isn't this sleight of hand? It's possibly true for the firm, as higher productivity allows lower costs, and greater market share. But at the sector level, or economy level, it is surely rather more dubious.

If we look at the manufacturing sector over the very long run, it's productivity has increased vastly, but employment certainly has not. Demand for manufactured goods simply isn't elastic enough to absorb the additional output.

At the economy level, rising productivity and rising employment will only go hand in hand if you make assumptions about what is happening to aggregate output, or (equivalently) if you assume nice neoclassical labour markets. Which begs the question.

Or am I missing something ?

[Luis Enrique]

that paper also rather overstates the extent to which mainstream economics neglects things like labour utilization. for example, see these slides by the very mainstream Peter Klenow:

http://klenow.com/KlenowSED.pdf

[Glenn]

Chris I will look at the paper but it states the bleedin' obvious really.

I think that UK and OECD policy in general has emphasised productivity growth as the route to enhancing / maintaining long run comparative advantage.

The criticism of the UK is that its low productivity performance is the basis for long run economic calamity and declining performance.

In particular the UK is characterised by many sectors/activities being low productivity compared to international peers (such as the LSE CIP work on long tail of low productivity businesses in most industries). HM Treasury in the 1990s ascribed a lot of this down to lower investment levels.

In addition there is the nature of the UK economy historically. Low savings, high consumer expenditure tends to favour an economic structure of services, retail etc which tends to be lower productivity. Plus there's the nature of retail with fewer 'big box' stores compared to say, France - which lowers productivity. Also interesting how UK compares to the German model of high saving, low spending, and focus on high value added manufactures.

Sure an economy which, to generate higher levels of output, uses more workers - is good for generating jobs. But this has implications for the real income of households if it makes goods and services more expensive. It also has implications for long run comparative advantage.

The dual approach of the UK having low tech services and high tech services/manufacturing also speeds up the 'hollowing out' of the middle of the occupational labour market - i.e. disappearance of skilled trades.

[willem le roux]

Companies hire people that are productive, and fire people that are not. This behaviour should result in unemployed people being, on average, less capable than people with jobs. One would expect productivity per worker (and therefore, per hour worked) to be lower in countries with low rates of unemployment, like the US and UK, than in countries with historically higher rates of unemployment like France. Not sure if this effect has been measured or how significant it is, but I would guess it explains at least part of the productivity growth differential between French and British workers between 1980 - 2005.