The TUC measure is not however a measure of unemployment. It is based on a definition adopted by the US Bureau of Labor Statistics (BLS) called “U6” which provides a measure of labour market slack. This includes those unemployed by ILO definitions, but adds some people who are classified as economically inactive by ILO definitions (not in work or ILO unemployed) who say they want a job and people who are working part time but who say they would like to work full time.

The TUC has produced a fascinating chart that shows how the two measures have moved over time, starting in April 1993, the first quarter such estimates can be provided from UK data.  This suggests a strong structural element in the wider measure – in other words, high levels of labour market slack have been a semi-permanent feature of the UK labour market over the past 20 years and are not just a creation of the recession.  Recent trends show ILO unemployment increasing much faster than the wider U6 measure.

The TUC analysis has had to use UK data, not all of which quite corresponds with US data and concepts. The biggest difference is that the US measures only includes the inactive who want to work if they have looked for work in the past year and could start a job in the week of the survey. The TUC uses the only readily available figure available, the total number of inactive who say they want a job. However, a closer match with the American “U6” definition would significantly reduce this total and therefore the overall estimate of labour market slack in the UK.

When discussing these issues some commentators also refer to the number of involuntary temporary workers – people in temporary work who said they would like a permanent job. While these people’s labour market position is by no means as secure as those in a permanent full-time jobs they are not included in the US U6 definition or in the TUC total, as they are not strictly speaking an indicator of labour market slack – people are working at the time of the survey even if they would prefer a more secure job.

More plausibly taken into account are the large number of part time workers who said they would like a full time job.  The US measure includes all such part time workers, and so does the TUC. However, a stricter measure of labour market slack would try and measure the difference in hours that people are actually working against the hours they would prefer. Average hours actually worked by part time workers in the UK in their main job are about 40 per cent of average actual hours worked by full time workers.

These measures of labour market slack are useful not just to highlight the jobs challenge facing any government, but also to give analysts and policy makers a better idea of just how much labour is readily available to sustain a recovery.  In recent years both the Office of Budget Responsibility and the Bank of England have looked at wider labour market slack measures for exactly this reason. The TUC measure is not a perfect match for the U6 indicator, but it can be refined and developed. Even more useful would be the adoption of a similar measure by the Office for National Statistics based on the U6 definition used by the US Bureau of Labor Statistics and published alongside the ILO unemployment measure.

The most recent growth figures (GDP) are a reminder that economic forecasting is a mug’s game. They show the UK economy was still contracting in the third quarter of 2009. Taking other indicators into account, we are still likely to be at a turning point and the numbers may well be revised upwards later this year.

But it is a salutary reminder of how easily hopes of an early recovery can be dashed. We have even less idea of the shape of the recovery – whether it will be a V, W or some other letter. The longer-term impact of the credit crunch on investment and consumer spending and the effectiveness of measures to free up the supply of credit – including “quantitative easing” – are still unknown. The only thing we can be fairly sure of is that the recovery in output to pre-recession levels will take no more than 4 years (a similar period to the Great Depression in the 1930s).

At the start of the recession much of the media was obsessed with the idea that this was going to be a middle class recession, centred on the finance sector in London. People leaving Canary Wharf with cardboard boxes made for dramatic pictures and TV crews prowled the streets of Surbiton, looking for evidence of recession. This was nonsense then and now. The recession is following a familiar path – it is the young, manual workers, administrative and sales persons, and the less skilled who are worst hit. And it is the cities and towns outside London in North Britain and the Midlands and Wales with low skilled populations, which have seen the biggest rises in unemployment.

The labour market has also performed better. Employers have been “labour hoarding”  – so that even though the fall in GDP looks similar to the recession of the 1980s and much deeper than the recession of the 1990s, far fewer people have been laid off. Equally encouraging, the surge in long-term worklessness that we saw in the last recession has not so far come through. Both suggest we could see a faster labour recovery than in previous downturns.

This is still a big “if” – in labour market terms we are still in the early stages of the recession and long-term problems emerge towards the end, not the start of recessions. Historical experience tells us that labour markets take between 8 and 10 years to recover to their pre-recession levels of employment. Even an optimistic view tells us it will be many years before employment fully recovers.

If forecasting is a mug’s game, how much can we say about where the new jobs will come from? As part of The Work Foundation’s Knowledge Economy Programme we have looked at what happened to employment in the decade after the start of the 1980s and 1990s recession to help us frame what a sustainable knowledge based economy might look like in 2020.

So where will growth come from in this recovery?

One source is overseas. The exchange rate fall is even greater than in the early 1990s and in the 1990s recovery UK manufacturing experienced a revival. Traded goods and services – including most of the manufacturing sector – will be more competitive. But much depends where the demand overseas recoveries fastest. The UK is less well positioned to benefit from faster growth in India and China than in the US and the EU.

Some manufacturing jobs may come back, but not on the production side. Manufacturing is transforming itself to a sector which combines high value services with some UK based production. It is in services associated with the manufacturing process that we will see growth. The net contribution of manufacturing to jobs growth is likely to be very small, but the sector’s contribution to the growth of value added, exports (including services) and innovation could be substantial.

The second is in the growth of high value service industries in the private sector such as high tech, business and communications services. These industries have provided much of the growth in jobs and value added over the past 30 years and helped lead the economic recoveries of the 80s and the 90s. These industries do not just provide jobs for graduates – the majority of jobs are held by non-graduates across a range of skill levels. But these jobs require different skill-sets to those being lost – these industries typically employ few manual and unskilled workers and routine admin jobs are being phased out by new technologies.

The third area is the most problematic. In both the 80s and 90s recoveries there was significant growth in both healthcare and education services, even those these periods were also years of public sector austerity. Over the next decade demand for these services is likely to grow while the ability of the public sector to supply them is going to be highly constrained. How much of the growth in education and health was in the private sector and/or funded by non-taxpayer sources in previous recoveries is surprisingly hard to quantify, but knowing the answer may give us some clue on the potential for growth in these areas in this recovery.

The fourth area is also difficult to predict. Traditional services such as hospitality and retailing can contribute large numbers of jobs. But retailing may no longer be the big job generator as new technologies and shifts in shopping towards on-line markets will reduce the labour intensity of the sector.

And what is completely unknown is how strongly consumer demand will return. Households always save more in downturns, so the current increase in the “savings ratio” (the share of household income saved) tells us nothing about the longer term. But if there is a structural shift towards permanently higher levels of saving above that in previous recoveries, then the longer run prospects for growth in traditional service sectors is correspondingly reduced.

Finally, the financial services sector can’t be written off entirely. In the UK at least, it has not been a big net generator of new jobs over the decade before the recession and is clearly contracting at present. But just suppose the big banks were broken up, new players outside the banking sector enter, and we give full encouragement to the rise of new financial institutions at the local and regional level. Perhaps, just perhaps, the sector could become a source of sustainable growth and jobs in the future.