This is a revised version- comments on the previous one made me realise that in trying to be brief I had instead just been confusing.

Productivity is usually measured as the value of the output produced by a worker in an hour. At national level, it is total GDP divided by the number of hours required to produce it. For a manufacturing economy this is conceptually easy to visualise (though not easy to measure). It conjures up comparisons between highly productive German workers with lots of shiny machinery producing more widgets per hour than British workers with older machines and fewer of them.

The problem is that we are now largely a service sector economy, where the concept of output per hour is harder to measure, the meaning of productivity is less clear, and the social benefits of alleged ‘productivity improvement’ can be ambiguous .

In the public sector, output is not sold and ( for international comparisons at least) we simply value the output according to how much it costs. If Government spends more but employ no more labour hours, then measured productivity increases. Of course output may have increased, perhaps as a result of spending on improving technology , but it could equally be the case that Government simply paid the cost of a wage increase and nothing changed at all, other than happier workers.

Productivity is also a tricky concept in the private sector and in the murky grey areas of public financing but private provision. If Government increases the minimum wage, for example, the outcome depends on whether firms are able to raise their prices. If they raise their prices by the full amount, then measured productivity will increase even if the same workers are delivering the same number of packages, or the same care workers are providing the same level of care to the same number of people. Higher prices raise the recorded value of output, and will be measured as a productivity improvement, even if the same workers are doing the same job in exactly the same way.

In practice the price increase is not the only change that will happen. There might be some fall in demand because of the price increase- which may mean some people losing their jobs because the enterprise they work for cuts costs or goes out of business. The loss of the least productive firms will raise the average productivity of the sector . There might be some increase in the actual output per hour as firms cut staff and remaining workers are made to take ever shorter breaks and given more challenging targets. Both the job cuts and the increased pressure to make those who stay work harder will be measured in the data as improvements in productivity, though neither is necessarily desirable. The firms that remain may be those that survive by cutting corners and over-working their staff. Reducing standards in care homes to avoid going bust is not what people usually think of when asked to describe productivity improvement.

The key point I want to make is that low productivity work exists because workers are willing to work for low pay in poor conditions. They are willing to do that because they have no alternative that offers them a better standard of living. The low productivity work at the lower end of the income scale will cease to exist if the state provides a reasonable safety net through a combination of benefit payments and/or the enforcement of minimum wages and conditions.

In the Labour years 1997-2007, the UK productivity growth measured by output per hour worked was second only to the USA among the leading economies of the G7. This was the period of increased minimum wages and improvements in the social safety net that dramatically reduced poverty. In the subsequent period of Tory Government, only Italy had slower productivity growth. This was the period of austerity when that safety net was allowed to collapse. The deterioration in relative productivity growth in the latter period is not an accident. It is a consequence.

The low level of unemployment in the UK is also a consequence of our inadequate safety net. As I have argued elsewhere in this blog, unemployment is a luxury good. Rich countries pay people to be idle because they are not willing to tolerate the levels of poverty that result if people are forced to subsist on the income from poorly paid jobs in harsh conditions.

Summarising, low productivity jobs are only profitable for employers if the cost of employing someone is lower than the value of what the worker produces. Stagnant productivity growth is a by-product of low wages, which are in turn made possible by the lack of a safety net that would have given workers a viable alternative.

If the Government is serious about driving higher economic growth based on improving productivity, then a good place to start is by improving the welfare safety net, raising the minimum wage, improving employment conditions . Average productivity will rise . Unemployment may also rise as workers are no longer forced into jobs at poverty wages.

Closing the option of making high profits by paying low wages to workers who have few choices does not necessarily result in firms deciding to invest instead in the technology of the future- though it probably makes it more likely. The growth agenda still needs support- as I argued in the previous post this is about supporting trade openness and improved education and training. However, improvement in productivity that is meaningful for the majority of the workforce also needs to encompass improvements to the safety net and to working conditions-the exact opposite course to the one the Tories are embarked on.