Spatially concentrated deprivation in England: an empirical assessment Alasdair Rae To cite this version: Alasdair Rae. Spatially concentrated deprivation in England: an empirical assessment. Regional Studies, 2011, pp.1. ￿10.1080/00343404.2011.565321￿. ￿hal-00694716￿ HAL Id: hal-00694716 https://hal.science/hal-00694716 Submitted on 6 May 2012 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Regional Studies F o r P e e r R Spatially concentrated deprivation in England: an empirical assessment e v i Journal: Regional Studies e Manuscript ID: CRES-2010-0335.R1 Manuscript Type: Main Section w J0 - General < J - Labor and Demographic Economics, J40 - General < J4 - Particular Labor Markets < J - Labor and Demographic Economics, J60 - General < J6 - MoObility, Unemployment, and JEL codes: Vacancies < J - Labor and Demographic Economics, J64 - Unemployment: Models, Duration, Incidennce, and Job Search < J6 - Mobility, Unemployment, and Vacancies < J -l Labor and Demographic Economics y deprivation, poverty, disadvantage, spatial, concentration, labour Keywords: market area http://mc.manuscriptcentral.com/cres Email: [email protected] Page 1 of 51 Regional Studies 1 2 3 Spatially concentrated deprivation in England: an empirical assessment 4 5 6 7 8 Alasdair Rae 9 Department of Town and Regional Planning 10 University of Sheffield 11 Sheffield 12 13 S10 2TN 14 United Kingdom 15 16 [email protected] 17 18 o 19 20 r 21 22 23 P (Received October 2010: in revised form February 2011) 24 25 e 26 e 27 28 r 29 30 31 R 32 33 e 34 35 v 36 i 37 38 e 39 40 w 41 42 43 O 44 45 46 n 47 l 48 y 49 50 51 52 53 54 55 56 57 58 59 60 1 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies Page 2 of 51 1 2 3 4 5 Abstract 6 7 8 Spatially concentrated deprivation is a well documented phenomenon and is of 9 10 interest to a diverse constituency of academics and policy makers. Despite the 11 12 13 accepted view of concentrated deprivation as a problem, however, the empirical 14 15 basis for understanding it remains under-developed. Therefore, an attempt is made 16 F 17 here to provide an empirical assessment of spatially concentrated deprivation in 18 o 19 20 England, using spratial statistics and a policy-relevant deprivation measure. More 21 22 localised analyses are also conducted for London, Birmingham, Liverpool and 23 P 24 25 Manchester. The resultes demonstrate that deprivation in England is highly 26 e 27 concentrated, that it varies significantly over space and that spatial patterns persist 28 r 29 through time. 30 31 R 32 33 e 34 Keywords: deprivation, poverty, disadvantage, spatial, concentration, labour market 35 v 36 i area 37 38 e 39 40 w 41 JEL Codes: J0 - General < J - Labour and Demograp hic Economics, J40 - General 42 43 < J4 - Particular Labour Markets < J - Labour and DOemographic Economics, J60 - 44 45 46 General < J6 - Mobility, Unemployment, and Vacanncies < J - Labour and 47 l 48 Demographic Economics, J64 - Unemployment: Models, Duration, Incidence, and y 49 50 Job Search < J6 - Mobility, Unemployment, and Vacancies < J - Labour and 51 52 53 Demographic Economics 54 55 56 57 58 59 60 2 http://mc.manuscriptcentral.com/cres Email: [email protected] Page 3 of 51 Regional Studies 1 2 3 1. Introduction 4 5 The phenomenon of spatially concentrated deprivation has interested policy 6 7 8 makers and academics for many years and has resulted in a wide variety of different 9 10 interventions which attempt to combat its perceived ill-effects. The reason why 11 12 13 spatially concentrated deprivation is seen to be problematic is often linked to 14 15 arguments for the existence of ‘area’ or ‘neighbourhood’ effects (GALSTER et al., 16 F 17 2007), whereby the impact of larger areas of poverty are seen to adversely affect life 18 o 19 20 chances above anrd beyond individual characteristics (SAMPSON, 2003). Despite 21 22 seemingly powerful arguments against the importance, or strength, of the 23 P 24 25 neighbourhood effect (ee.g. CHESHIRE, 2007; VAN HAM and MANLEY, 2010), 26 e 27 however, it has become commonplace for governments in many nations to employ 28 r 29 social mix strategies in order to break up (or dilute) large concentrations of 30 31 R deprivation, poverty or disadvantage (e.g. GALSTER, 2007). Beyond the 32 33 e 34 neighbourhood effects issue, there are of course more fundamental questions of 35 v 36 i spatial justice associated with concentrations of deprivation (DABINETT, 2010). 37 38 e 39 Notwithstanding wider concerns about why policy focuses almost exclusively on 40 w 41 concentrated poverty rather than, or as well as, con centrated affluence (MASSEY, 42 43 1996; COULTON et al., 1996) there remains a lack ofO empirical evidence in relation 44 45 46 to the scale of the phenomenon and the extent to which thne spatial concentration of 47 l 48 deprivation may change over time. Even where commentators disagree about the y 49 50 importance of the neighbourhood effect on individual outcomes, the existence of 51 52 53 concentrated deprivation, particularly in inner-urban locations, is not debated. With 54 55 very few exceptions (e.g. GALSTER, 2005; DORLING et al., 2007; OAKLEY and 56 57 58 LOGAN, 2007), an empirical basis for understanding this concept is lacking. Within 59 60 the contemporary ‘evidence-based’ policy paradigm it is non-existent. This is an 3 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies Page 4 of 51 1 2 3 important shortcoming, since policies which seek to foster a more mixed society 4 5 involve significant public expenditure and have enjoyed widespread support. If there 6 7 8 is no real objective basis for measuring concentration then the kinds of policies 9 10 currently and recently in operation cannot, in the contemporary fiscal climate, be 11 12 13 easily justified – even if they appear to be effective. This research therefore seeks to 14 15 take a step back from wider debates about social mix strategies and neighbourhood 16 F 17 effects (e.g. DARCY, 2007; LUPTON and FULLER, 2009) in order to provide the 18 o 19 20 kind of spatial interlligence which can help policy makers and academics understand 21 22 the geography, intensity and persistence of the spatial patterns which motivate 23 P 24 25 intervention in the first plaece. This is done by presenting a series of empirical findings 26 e 27 on patterns of concentrated deprivation in England. 28 r 29 There are four sections in this paper, followed by some concluding remarks. 30 31 R The first section briefly introduces and explores the theoretical lineage and 32 33 e 34 contemporary policy context behind the concept of spatially concentrated deprivation 35 v 36 i in order to foreground the subsequent empirical analysis. The second section 37 38 e 39 considers the analytical background to the analyses which follow and provides 40 w 41 further details of the specific dataset used. The third section reports on the empirical 42 43 research undertaken. It presents a series of more inO-depth findings relating to the 44 45 46 nature of spatially concentrated deprivation in England on ann annual basis from 1999 47 l 48 to 2005 – an important time period in the recent history of English urban policy. This y 49 50 is first done at a national level, with a subsequent focus on four key urban areas. The 51 52 53 evidence presented here is based on comprehensive spatial statistical analysis. A 54 55 fourth section then attempts to draw together the findings and focus on key 56 57 58 messages emerging from the foregoing empirical section. 59 60 4 http://mc.manuscriptcentral.com/cres Email: [email protected] Page 5 of 51 Regional Studies 1 2 3 2. Understanding Spatially Concentrated Deprivation 4 5 In addition to concentrated deprivation, socio-spatial inequalities are often 6 7 8 approached from two related perspectives, with some authors focusing on 9 10 concentrated ‘disadvantage’ (e.g. CARPIANO et al., 2009) and others on 11 12 13 concentrated ‘poverty’ (e.g. LUPTON and FULLER, 2009). Although these concepts 14 15 overlap in many aspects and distinctions are sometimes difficult to make, there are 16 F 17 important definitional differences which mean they cannot be treated as entirely 18 o 19 20 synonymous in anr empirical study – even if policy responses to each are similar. 21 22 Spatially concentrated poverty is typically operationalised in relation to the 23 P 24 25 percentage of people in aen area who are in ‘poverty’, according to some nationally 26 e 27 contextualised median household income value (e.g. GALSTER, 2005). Spatially 28 r 29 concentrated disadvantage, on the other hand, is a more relative concept, which can 30 31 R be applied to different groups in different ways (e.g. NATION, 2008) and can also be 32 33 e 34 extended to incorporate the more generic construct of ‘locational disadvantage’ 35 v 36 i (DARCY, 2007). The term ‘deprivation’ is adopted here since the data used are part 37 38 e 39 of an official deprivation index relating to ‘economic deprivation’ in England (CLG, 40 w 41 2009) and the results relate specifically to spatial co ncentrations in this sense. The 42 43 initial sections of this paper recognise the overlaps betOween deprivation, poverty and 44 45 46 disadvantage – particularly in relation to policy responsesn – whereas the empirical 47 l 48 sections relate specifically to the concept of ‘deprivation’ as defined by the Economic y 49 50 Deprivation Index (CLG, 2009). 51 52 53 The spatial clustering of poverty, disadvantage or deprivation is not a new 54 55 phenomenon; a fact noted by numerous commentators (e.g. JONES, 1979; 56 57 58 GALSTER, 2005; NORTH and SYRETT, 2008). Concomitantly, neither is the idea 59 60 that such clustering is a problem and that more balanced communities are a solution 5 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies Page 6 of 51 1 2 3 (e.g. LUPTON and FULLER, 2009). Nonetheless, over the past fifteen years in 4 5 particular there has been something of a re-invigoration of attempts to do something 6 7 8 about spatially concentrated deprivation, and its presumed pernicious effects. This 9 10 has come to fruition in a number of countries and by means of a variety of quite 11 12 13 different urban policy interventions. Examples include the much-cited ‘Moving to 14 15 Opportunity’ experiment in the United States (GOERING and FEINS, 2003) and the 16 F 17 more broad-brush ‘Big Cities Policy’ in the Netherlands (VAN KEMPEN and BOLT, 18 o 19 20 2009). In Englandr, the New Deal for Communities programme (LAWLESS, 2004) 21 22 and its contemporaries were symbolic of the New Labour ‘urban renaissance’ project 23 P 24 25 and an ideological commeitment to a form of spatial and social justice. The literature 26 e 27 on this topic is now voluminous and rightly critical since the results now seem at 28 r 29 odds with the rhetoric. The intention here, though, is to consider how spatial 30 31 R concentration has been approached and understood in relation to the associated 32 33 e 34 phenomena of deprivation, poverty and disadvantage, and to highlight the lack of 35 v 36 i empirical evidence in relation to the precise nature of concentrated deprivation as it 37 38 e 39 is currently conceptualised. 40 w 41 One of the less well noted features of rece nt literature on concentrated 42 43 deprivation, poverty or disadvantage is the extent to wOhich the phenomena remain, 44 45 46 in spatial empirics, somewhat ill-defined. This is in contranst to their identification as 47 l 48 phenomena per se (e.g. SAMPSON et al., 1997; COHEN et al., 2003; NATION, y 49 50 2008; CARPIANO et al., 2009). Thus, important contributors such as NORTH and 51 52 53 SYRETT (2008) accurately describe the problem of concentrations of deprivation 54 55 with reference to spatial patterning across England at a moment in time, but do not 56 57 58 measure it in a spatial statistical manner. This is not a problem in itself, of course, 59 60 but if we are to understand the impacts of policy intervention on concentrated 6 http://mc.manuscriptcentral.com/cres Email: [email protected] Page 7 of 51 Regional Studies 1 2 3 deprivation through time and space it is useful to have some reference point from 4 5 which to begin, hence this research. First, however, it is important to have some 6 7 8 understanding of how it has been conceptualised and how this has fed into the policy 9 10 mainstream, if at all. 11 12 13 In the wide-ranging literature that spans the social sciences, there are many 14 15 different perspectives on the problem of spatial concentration; this can be seen in 16 F 17 relation to studies on deprivation (RAE, 2009), poverty (GALSTER, 2005) and 18 o 19 20 disadvantage (CArRPIANO et al., 2009). However, it is possible to identify four 21 22 relevant, inter-related themes here. First, we have the fundamental capitalistic 23 P 24 25 uneven development theesis, propagated most powerfully by David Harvey (e.g. 26 e 27 HARVEY, 1985), which views the geographical nature of deprivation as an artefact 28 r 29 of the spatial manifestation of capitalism. In this view, spatial concentration is a 30 31 R symptom rather than a cause of social disadvantage. Related to this ‘concentration 32 33 e 34 as symptom’ stance, there is a second view of concentration as something of a 35 v 36 i distraction in wider debates on poverty and place. In this view, commentators have 37 38 e 39 either highlighted the inherent limitations of contemporary policy (CHATTERTON 40 w 41 and BRADLEY, 2000) or highlighted a lack of jus tification for mixed community 42 43 policies in the first place (CHESHIRE, 2007). Very Odifferent intellectual traditions 44 45 46 here lead to similar conclusions (cf. LUPTON and FULLERn, 2009). Third, there is the 47 l 48 view of spatial concentration as a remnant of historical-geographical processes y 49 50 which consume space differently at different times (cf. MASSEY, 1979). In this 51 52 53 perspective, it is longer-term macroeconomic and macro-spatial forces which are 54 55 given primacy, and policy solutions are not necessarily spatial. A fourth element 56 57 58 emerging from the literature is the tendency to focus on spatially concentrated 59 60 deprivation, poverty or disadvantage as isolated phenomena so that concentrated 7 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies Page 8 of 51 1 2 3 affluence largely remains unchallenged (see MASSEY, 1996; SAMPSON et al., 4 5 2002). This remains one of the most surprising aspects of the subject matter; 6 7 8 particularly in disciplines which contain so many critical voices. Given these factors, 9 10 and the potency of their central arguments, one might expect policy responses to 11 12 13 socio-spatial inequalities to encompass a wide range of different measures, including 14 15 macro-economic stimulus and public transport investment. Instead, the promotion of 16 F 17 ‘social mix’ strategies (e.g. ARTHURSON, 2002) has dominated and one is left to 18 o 19 20 question if there isr actually a serious disconnect between theory and practice here. 21 22 In his paper on the need for a more reflexive epistemology in spatial social 23 P 24 25 science, DARCY (2007) neotes that in contemporary public housing policy discourse, 26 e 27 ‘concentration of disadvantage’ has been identified as the problem and ‘social mix’ 28 r 29 presented as the solution. Similarly, GALSTER (2007) comments on the adoption of 30 31 R such strategies by several western European countries, and DUYVENDAK et al. 32 33 e 34 (2009) comment on the commitment to, but differences in, social mix strategies 35 v 36 i between the US and France on the one hand and the Netherlands and UK on the 37 38 e 39 other. Given this explicit focus on concentrated deprivation, poverty or disadvantage 40 w 41 as a problem, and the culture of evidence-based pol icy which predominates, then, it 42 43 is reasonable to expect that any policies implementeOd would have some basis for 44 45 46 assessing empirically the extent, nature and trajectory of nspatial concentration. But 47 l 48 this is not the case. There have been studies which explore this issue in various y 49 50 ways (e.g. DORLING et al., 2007; PATACCHINI and RICE, 2007) and attempts to 51 52 53 ‘spatialise’ existing attributional measures of deprivation at a local level (RAE, 2009), 54 55 but spatially concentrated deprivation remains a somewhat opaque concept in a 56 57 58 formal sense. We know that it exists, that it is important and that it persists through 59 60 time, yet we do not formally assess its intensity, its trajectory or its scale. Given the 8 http://mc.manuscriptcentral.com/cres Email: [email protected]