Getting worklessness wrong, again

‘Worklessness’ is one of those terms which means one thing in specialist usage and something quite different in political discourse and media commentary.

‘Worklessness’ is one of those terms which means one thing in specialist usage and something quite different in political discourse and media commentary. In labour market statistics a workless household is a household including one or more working age adults in which none of the adults is working (at a particular point in time, or over a period if the data tracks people over time).

This contrasts with what the Office for National Statistics refers to as ‘working households’ where all working age adults are in employment (including households with only one adult) and ‘mixed’ households in which some are working and some aren’t. It is a descriptive term for a situation, not a term for a personal characteristic of individuals or their households or families.

In political discourse, a workless household is something quite different, a loosely constructed amalgam of moral intuition and vague notions of social dysfunction and dependency extending across lifetimes and even across generations. Will Hutton wrote in yesterday’s Observer:

The welfare state was not set up to support vast families or single mothers in intergenerational welfare dependency.

That is not a bad summary of what ‘workless household’ has come to mean in contemporary discourse.

The fact that a term is used to refer to different concepts in different contexts would be less of a problem if evidence concerning the one was not routinely used as if it concerned the other. Moreover, when it comes to inappropriate use of evidence, the traffic is one way: it is data on household worklessness in the first, descriptive sense which is cited to provide spurious statistical support for statements about worklessness in the second sense, and not the other way round.

Thus many of the statements we hear from politicians and commentators about intergenerational or long term worklessness and ‘welfare dependency’ are parasitical on data which does not concern these phenomena, as has been pointed out here before.

Household worklessness, and more generally household employment patterns, are serious issues which need to be approached with due respect for the evidence. Today’s worklessness stories turn on a report from the Centre for Policy Studies which was press released ‘UK HAS MORE WORKLESS HOUSEHOLDS THAN OTHER MAJOR EU COUNTRIES’.

There is no great problem with this claim: it is true that the Eurostat data in question relates to ‘working age people in workless households’ rather than ‘households’, which is a different concept and a different figure, so the headline is misleading; and it is also true that the report for some reason omits Belgium, which has had a higher percentage of people in workless households than the UK for many years, from its comparative table.

But it is the case that the UK has long had one of the highest rates in Europe, even if its position relative to other European countries has moved up and down over the years: for example, Germany and France have both had rates very close to the UK within the last five years. The average for the EU 15 was 10.1% in 2009 and the UK had a rate of 11.5%. (That is a lower figure than that given by ONS in its recent report, which shows a rate of 13.5%; however, the latter figure cannot be compared with the Eurostat data used by the authors.)

What is extraordinary is that the authors choose to contextualise this data in the following way:

In 1979, the UK economy was invigorated by the privatisation and liberalisation of nationalised industries, public corporations and local authority housing. In 2010, it is socialised households that require “privatisation”.

Here is the history of how the invigoration of the UK economy in the 1980s played out for households. In 1977, 5.8% of individuals in the UK were living in workless households: that rate rose to 11.8% in 1986, fell during the Lawson boom but only to 9.8% and rose again to 14.2% in 1996. Since then the rate has fallen gradually, reaching a low of 12.3% in 2008. With the onset of recession the rate has risen to 13.5%. That rise underlines the point that the workless rate is not to be confused with long term economic inactivity. (Figures from Gregg and Wadsworth; see here).

But there is no question that the 1980s saw a structural shift in household employment patterns, some of it explicable in terms of household structure and demographics but much of it driven by labour market changes and government policy: the latter including a willingness to use incapacity benefit as a form of subsidised early retirement for older industrial workers, thus reducing headline unemployment, and a stance of at best benign neglect of lone parents, whose employment rates fell compared to the late 70s and only began to rise consistently when Labour made them a priority from 1997.

That the 1980s saw devastating changes in household employment, poverty and on social attitudes towards inequality which have yet to be reversed is undeniable. At the same time, it is a mistake to infer from headline worklessness rates that nothing has changed in the last 20 years. Headline figures are driven by factors which can push in opposite directions.

The growth of single person households has had a major impact on the number and rate of workless households in the UK, acting as an offset to falling rates of worklessness across household types: while there is no reason to be relaxed about workless people living alone, they do not fit the social stereotypes associated with the workless household.

Sharp falls in incapacity benefit receipt among men over fifty have been offset both by demographic change (there are more men over fifty, even if they are less likely to be on ICB) and by increasing numbers of claims for mental health related conditions among younger people (some but not all of which will reflect changing policy approaches to mental health over the longer term).

The rate of youth unemployment, which the authors attempt to link without evidence to the household worklessness figures, is driven not just by the numbers of young people not in education, employment or training but by rising numbers in education (the denominator is economically active young people, so students are excluded, pushing up the unemployment rate). Couples with dependent children are more likely to be workless now than they were two years ago (5.4%) but a lot less likely than in 1997 (6.9%).

The most unambiguous positive change has been in lone parent employment. The chart below shows how employment rates have changed for lone parents, mothers in couples and people without children since 1997, and points to two important aspects of developments over the last 13 years.

The first is that the shift in the employment rate is structural, not just the reflection of a booming labour market over most of Labour’s period in office. The recession has had remarkably little impact on lone parent employment (and on the employment of mothers in couples), especially when compared with those without children.

The second is that the growth of lone parent employment was more marked during Labour’s first two terms in office. Policy is not the only factor in explaining this change, but it is one of the factors, and it is striking that the policies which contributed to growth in lone parent employment up to 2005 included tax credits, the minimum wage, Sure Start and mild conditionality but did not include benefit cuts and the other ‘get-tough’ policies which seem now to be regarded as the only relevant policy instruments.

Employment-by-parental-status-1997-2010

A report pointing to the economic and social advantages that can arise from a focus on the relationship between households and labour markets ought to be welcome – especially one that stresses that economic growth alone does not guarantee high levels of household employment and questions the consumption based patterns of growth the UK has experienced for so long. But references to the “structural problem of worklessness” need to be backed up with an understanding of stability and change over the last 30 years which is lacking here.

The CPS report glosses over the impact of policies pursued during the 1980s on household employment, argues in all seriousness that tax credits are a cause  of worklessness and recommends a series of off-the-peg policy solutions which show no serious engagement with the last 13 years of policy development in this area, but which are sure to appeal to those for whom worklessness is seen as a homogenous intergenerationally transmitted characteristic demanding disciplinary action rather than as a range of different situations, arising from a host of factors and requiring attention to all relevant policy instruments.

There seems to be an intense demand for simple stories about worklessness at the moment – and a seemingly limitless supply, if somewhat lacking in variety. An intellectually honest approach would start by telling audiences from all parts of the political spectrum that this is an area where if a story is simple, it is very unlikely to be true.

12 Responses to “Getting worklessness wrong, again”

  1. paulstpancras

    RT @leftfootfwd: Getting worklessness wrong, again: http://bit.ly/agN5Zc – Declan Gaffney on what the worklessness stats actually mean

  2. R Gordon

    RT @paulstpancras: RT @leftfootfwd: Getting worklessness wrong, again: http://bit.ly/agN5Zc – what the worklessness stats actually mean

  3. Lynn Hancock

    RT @leftfootfwd: Getting worklessness wrong, again http://bit.ly/agN5Zc – great analysis here.

  4. Alison McGovern

    RT @leftfootfwd: Getting worklessness wrong, again: http://bit.ly/agN5Zc – Declan Gaffney on what the worklessness stats actually mean

  5. yorkierosie

    RT @leftfootfwd: Getting worklessness wrong, again http://bit.ly/agN5Zc

Comments are closed.