Structure rather than behaviour: on the causes of poverty

Structure rather than behaviour: on the causes of poverty
London School of Economics, by Glen Bramley 


Whether poverty is caused by behaviour or economic factors is an ongoing debate, with the Right usually supporting the former, and the Left the latter. Drawing on a recent study, Glen Bramley explains why structure, rather than behaviour, is the major factor behind disadvantaged individuals in England. Yet he also writes that many of those included in the research had adverse childhood experiences, and so the Left is also wrong to completely dismiss any suggestions of “transmitted poverty” between generations.


David Cameron’s recent speech on Life Chances attempted to reset the agenda on policy to tackle poverty and deprivation. Four main strands of policy were claimed to stem from four key insights from a more ‘social’ (dare one say ‘evidence based’?) approach to understanding poverty. The last of these was introduced in this way:


…when we know that so many of those in poverty have specific, treatable problems such as alcoholism, drug addiction, poor mental health we’ve got to offer the right support, including to those in crisis.


Probably not by coincidence, the very next day Ian Duncan Smith, Secretary of State for Work and Pensions, shocked readers of the Daily Mail with: ‘Addicts and alcoholics cost us £10billion a year, says Duncan Smith: Blitz launched to help people with drink drug problems find work’. The numbers in this statement came from a report we published one year ago, – Hard Edges: Mapping Severe and Multiple Disadvantage. Arguably this study shines fresh light on a centuries-old debate about whether poor people owe their circumstances to structural economic factors or to moral/behavioural failings. Since much of the emerging agenda appears to suggest, directly or implicitly, that the latter factors are more important than the former, it seems appropriate to highlight these findings.


Recent debates about the redefinition of poverty, instigated by Iain Duncan Smith, but echoing the contribution of Sir Keith Joseph in the 1980s, appear again to raise the profile of the second group of factors, while introducing a longitudinal inter-generational aspect (as in ‘bad parenting’). Stances on this debate fall mainly along predictable political/ideological lines, the Left favouring the structural and the Right favouring the personal accounts, with the academic discipline of social policy falling very heavily into the former camp.


The Hard Edges study shines a new and striking light on this long-running standoff. The report focusses on a group who may be considered the ‘poster-boys’ (they are mainly male, although not that young), for the moral/behavioural account, a group who sit on the ‘Hard Edges’ of society and social service provision. It attempts to provide the first comprehensive profile of adults suffering from ‘Severe and Multiple Disadvantage’ (SMD), namely combinations of homelessness, chronic offending and substance misuse. The study covers England, drawing on and triangulating evidence from national administrative systems and some targeted as well as more general surveys.


This group are only a small subset of the poor population (around a quarter of a million working age adults, or 4 per cent of the 7 million living in relative low income poverty). Thus, it is quite wrong to portray this as a major, dominant cause of poverty. But, with their complex needs and problems, this group generate quite high levels of financial and social costs on society, while themselves experiencing very poor outcomes and quality of life.


But the really interesting finding, from the viewpoint of the eternal social policy debate, is what emerges from the geographical crunching of the numbers. All three independent administrative datasets yield the same conclusion about the prevalence of severe and multiple disadvantage by local authority area. The top 10 areas are –wait for it – Blackpool, Middlesbrough, Liverpool, Rochdale, Manchester, Hull, Bournemouth, Nottingham, Stoke, Newcastle. With one exception, it is a roll-call of northern urban and industrial towns, the major sites of de-industrialisation and the highest concentrations of low income poverty. The bottom ten are southern affluent commuter suburbs and semi-rural areas – Wokingham, Central Bedfordshire, South Gloucestershire, Windsor & Maidenhead, East Riding, Buckinghamshire, Harrow, Richmond-on-Thames, Surrey, West Berkshire.  This is a clear signal, not just a ghostly image on a cartographic Turin shroud; the stigmata of structural material poverty show through quite clearly, and are confirmed by statistical analysis.


So, if structure matters even for the groups most stigmatised for their personal reprehensible moral and behavioural degeneracy, how much more influential must it be for the generality of poor people? A working age adult in Middlesbrough is ten times more likely than one in Central Bedfordshire to be in our ‘SMD2/3’ category (experiencing two or more of homelessness, substance misuse and/or offending). This evidence strongly suggests that persistent and widespread material poverty linked to structural labour market weakness generates processes which lead to high levels of disadvantage. They say that ‘correlation does not prove causation’, but in this case reverse causation – i.e. the notion that addiction problems brought about the deindustrialisation of northern towns and cities – is not a serious proposition.


A further remarkable set of findings give some further insight into these processes. The study highlights evidence from one survey and one administrative data system to show just how strong the relationship is between complex needs and certain background factors going back to childhood. Adults now experiencing severe multiple deprivation, particularly the most complex cases, had very high instances of adverse childhood experiences, ranging from 42 per cent ‘ran away’, 31 per cent left home at 16-17, 30 per cent not get on with family, 29 per cent parents violent, 29 per cent parents drug/alcohol problems, through 25 per cent abused, 18 per cent neglected, 17 per cent starved and 18 per cent in care. Only 15 per cent had experienced ‘none of the above’. In other words, it may be argued that the principal determinant of experiencing severe multiple deprivation is adverse childhood experiences, in conjunction with the structural poverty highlighted above. Thus, although this group clearly compound their (and our) problems by bad behavioural choices, they got to that position as adults from a horrendous experience as children, for which they cannot be held responsible.


The background data on education is also extremely interesting. 45-49 per cent of SMD3 people have no qualifications, which is now a relatively rare condition, around 3 per cent, in the younger adult population. Between 48 per cent and 59 per cent had significant/persistent attendance/truancy problems and 47 per cent were suspended from school. In other words, there is a glaring opportunity for early intervention in the education sector to capture and turn around these potential cases of costly SMD in adult life.


This point is reinforced by another striking finding. A majority of SMD adults (and 59 per cent of the most complex group) in the treatment database were parents of children, whether living with them or not, or had other child contact (e.g. through partner’s children). Taken in conjunction with another scary statistic, that 42-48 per cent of SMD offenders were domestic violence perpetrators, this suggests that the danger of today’s generation of SMD adults transmitting problems on through the next generation are extremely high.


So what does this tell us about the eternal debate in social policy?


The ‘Right’ are wrong to claim that much or most poverty is attributable to personal behaviour rather than to structural economic factors. The adult SMD group, for whom dysfunctional behaviour is most likely to be a major factor in their current malaise, is only a small proportion of overall poverty (around 4 per cent). Furthermore, the fact that there are 10 times more of them in Middlesbrough than Central Bedfordshire can only be explained by structural factors. And indeed it is clear that the dominant individual level factors implicated are adverse childhood experiences, for which they cannot be held responsible. However, the ‘Left’/social policy establishment are wrong to try to claim that current behaviour is never a factor in poverty for any group, when we can see that it clearly is a problem for many of these complex-need adults experiencing SMD. And they are wrong to completely dismiss accounts about ‘transmitted poverty or disadvantage’ between generations, when clearly there are strong influences from childhood on today’s adults with complex needs. The point is to strengthen the interventions at this key stage to avert such trauma and consequent transmission.