Download presentation
260

DISCUSSION PAPER SERIES DPS16.27 NOVEMBER 2016 How does early deprivation relate to later-life outcomes? A longitudinal analysis Ron DIRIS and Frank VANDENBROUCKE Public Economics Faculty of Business Diris* Vandenbroucke† Abstract Measures material are increasingly used as alternatives traditional poverty indicators. While there exists extensive literature focusing on the impact that growing up in a (financially) poor household has future success, little is known about how relates long-run outcomes. This study uses data from 1970 British Cohort Study assess relationship between outcomes adult life. We control for an set observable characteristics, further employ valueadded generalized sensitivity nature this relationship. find diverse outcome variables, but magnitude conditional relationships generally small. Immaterial indicators family quality show relatively stronger ties outcomes, especially with respect non-cognitive skills. Keywords: deprivation, poverty, disadvantage JEL Classification: I32, J13, J62 *Department Economics, Maastricht University, 6200 MD Maastricht, Netherlands, [email protected] (corresponding author) †University Amsterdam, KU Leuven University Antwerp. would like thank Erwin Ooghe, Brian Nolan, Geranda Notten, Kristof de Witte, participants APPAM conference London PE seminar their helpful comments. 1 Introduction Classifications or social exclusion have traditionally relied measures individual income. Material (MD) alternative indicator exclusion. refer list ’basic necessities’ households different domains The increasing use these reflects perception captures more than lack Although depends what perceived basket necessities at given point time, it essentially absolute measure poverty. contrasts commonly at-risk-of-poverty rate, which relative income positions within country.1 become popular international intertemporal comparisons. ‘Severe deprivation’ included target Europe 2020 strategy European Union (European Commission, 2010). However, contrast specifically related important success. analyzes confronted (BCS), follows total 17,000 individuals born Britain first week April 1970. BCS reports information child its parents birth contains follow-ups multiple ages both childhood life, until age 42. It provides possessions circumstances well vast range variables several progress extensively extent raw correlations driven by associations other determinants progress. Moreover, we value-added developed Imbens (2003) address selection bias establish whether likely causal remains. Using factor analysis, six 1The character should be interpreted nuance. Poverty can also character, least context one country, when threshold anchored time. based arbitrary choice base year (in defined). is, principle, possible construct countries. Notten Roelen (2012) show, constructing basis hazardous exercise. 2 estimated. two strands literature: studies analyzing (or complementary) background former group mainly focuses explaining mismatch being income-poor materially deprived,2 encompassing items.3 Advocates emphasize benefits over strictly income, conceptual view (income neglects circumstances, preferences risk factors) terms measurement (yearly volatile across time prone error, extremes distribution). shows into matters greatly success There strong children later example educational attainment income.4 Evidence adoption indicates variation families not solely due genes, therefore ‘family quality’ crucial importance children.5 still unclear specific aspects capture quality. Studies composite socio-economic status (SES) typically combine parental education, occupation, home and/or linked those outcomes.6 Brooks-Gunn Duncan (1997) provide overview focus relation conclude life (preschool school years) most strongly remains difficult empirically disentangle occu- 2See, e.g., Perry (2002); Whelan et al. (2004). 3Different methods elicit single MD, such prevalence weighting, principal component item response theory structural equation modeling, no consensus exists. For examples each approaches, see, Cappellari Jenkins (2006); Maˆıtre (2005); Tomlinson (2008). An provided Nolan (2010). 4See, Corak (2013) intergenerational transmission and, OECD (2015) 5See, Bjorklund Sacerdote (2008); Beckett (2006). ¨ 6See, Bradley Corwyn (2002) overview. 3 pation quality, neighbourhood rearing behavior, etc. Recent aimed uncover direct links Many role credit constraints attainment. type research finds short-term becomes limited, best, once factors achievement concludes permanent markedly liquidity (Heckman, 2000; Carneiro Heckman, 2003; Dearden al., 2004; Chevalier 2013). Still, (permanent) factors. Several exploited exogenous directly impact. example, Frijters (2005), using sibling fixed effects combination event German reunification, identifies low health, while Løken (2010), Norwegian oil boom shock, Other identify comparatively larger estimates substantially below simple suggest; Blanden Gregg (2004) (partially same (British) paper) Akee These results call question provision will lead substantial improvements prospects families. (2009) through evaluation EMA program, students weekly cash transfers attendance. program staying school, clear alleviation constraints, because reduce opportunity costs education. Overall, findings tend suggest correlation large part variables. led researchers argue largely immaterial (see, Heckman (2008)). explanation limited that, advocates often argue, 4 only imperfectly restrictions opportunities face. As such, meaningful analyze exclusion, either substitute complement Establishing existing emphasis put policy evaluation, policies targeted reducing deprivation. Identifying relations towards improve evaluations policies. In general, few key Filmer Pritchett (1999) exception, conducting macro-level they link differences wealth (measured presence basic facilities drinking water electricity) Relying rich micro-level data, current various measured Additionally, add addressing potentially confounding likelihood effects, providing comparison sample. paper organized follows. introduce theoretical considerations Section 2. describes methodological issues discussed 4. 5 presents empirical results. 6 discusses robustness analyses, 7 concludes. Theory 2.1 Defining section, discuss concept arise measuring constructs definition states “material refers inability afford consumption goods activities typical certain society irrespective people’s items” (OECD, 2007). words, concerns able ‘typical’ goods. major broad characterization be. considerable exact construction Virtually all incorporate items housing conditions. More elaborate include access healthy lifestyle Since aim broadest sense, since unexplored domains, analysis. ultimately want affects developmental process, child’s learning development (outside formal processes extra-curricular programs) additional domain. define deprivation: possessional health already suggested definition, ‘material’ aspect always adhered. aspects. make distinction ‘immaterial’ discussion reflected subdivision possession, housing, domain nature.7 ambiguous contain tangible tools intangible support. thereby divide sub-domain separately might alternatively thought cultural ‘capital’ thereof). 7One crime material, believe conceptually tied household’s living arrangements (which evidently captured domain), categorize under material. types goes beyond aspect, (often depending subjective interpretations) see them things everyone ‘should have’. Ermisch (2008) makes similar his parenting inequality labels ‘what buy’ versus do’. consider light bigger matter most, simultaneously recognize aware comparing interpreting 2.2 Measuring Another variable. Data availability inevitably determines some any application, criteria employed. First all, clearly goods, services society, case 1970s 1980s. concern ‘enrichment’ available share population. connotation leaves room interpretation. study, specify constraint half could seen rather loose constraint, analyses estimated limit higher prevalence. result affordability preferences. reason, comprise questions distinguish having personal preference. looks children, who bound own predominantly parents. ranks value necessity, never completely ‘irrespective preferences’, technically requires.8 made odd, situated high-crime neighbourhood. reasons, do preference our main conduct stage where order final potential assessing considered data. natural consequence difference ‘concise’ rely very dataset rich. Hence, although lower bounds fact inexhaustible relevant include, viewed upper explain studies. Study, adulthood (we label ‘cohort members’). baseline characteristics 17,196 individuals, 5, 10, 16, 26, 30, 34, 38 42.9 suffer amount observations, sample drops out fairly (86% 10 73% 26). waves 0 8The interpretation dependent adaptive feelings shame recognized e.g. Fusco (2011), addressed analyses. exception (2006), adapt Item Response approach correct differential reporting propensities item. 9The wave around 8 administered only, tests. 16 following (i.e. cohort members). school-level teachers principals waves. taken conditions verbal wide services. focal somewhat weighted specification. supplementary Each background. Parental categorical variable (seven categories eleven 16). follow McKnight assigning midpoint band. including obtained quali- fications, status, mental body mass index, satisfaction, gross net structure. four variables: reading highest qualification, income,10 general health. qualification self-reported If missing, impute next recent observation. apply increases rapidly observe avoid missing observation leads measure.11 non-missing years established trends then calculate average express rank 100 10To confusion member’s serves serve ‘adult income’, ‘parental income’ control. 11The mean values stable age. 9 test scores intelligence, math. Questionnaires carried skills well. sets allow self-esteem, locus person feels life) Rutter index behavioral problems. reported members, latter 16. Locus self-esteem Estimation 4.1 Measurement mentioned 2, separate health/nutritional subdivided sub-domain. rate 50%. inputs take affected (intermediate) dummy education aspiration levels desired level particular child, performs school. Similarly, exclude number friends visited same-aged peers assume choices child. subjective, carry section ‘ambiguous’ items. determine, domain, best fit relevance uniqueness measures). choose method explanatory power weighting weight assigned inverse sample). cases, includes ages. Being deprived of, TV sources overlap items, automatically ensures much receive weight, excluded altogether. All standardized zero standard deviation 1.12 assessment approaches purpose paper, report 6.6 completion. presented Table A1.13 4.2 model estimate OLS model: Yi = β0+β1P ossi+β2Housei+β3Neighi+β4Healthi+β5EduMi+β6EduIi+β7Soci+θX0 i+i (1) vector X0 birth, employment, 12The Cronbach’s alpha are: 0.801 0.700 0.640 0.554 0.545 eduational 0.447 13The defined unlikely (e.g. appliances), chosen priors expected ex ante. affect indirectly by, spend child-rearing. 11 style, complete Appendix A2. inclusion account into, outside controls effect operate spending tutoring classes). on, among households. When available, ages, father’s employment without controls, impacts parameter  Model represents classical error term. represent achievement, attainment, status. mechanisms progress.14 array cognition socio-emotional play mediating outcomes.15 4.3 Imputation waves, observations To ensure enough sample, values. observed 14We channels shaped. contrast, grows in. 15See, Almlund (2011) cognitive 12 imputation Woßmann ‘fundamental’ (labeled F) fundamental virtually observations. birth; gestational age, mother’s ethnicity, wedlock gender, was hospital-born family. M, (Mk ) (Mj ). regress Mj F coefficients regression Mk . Further, dummies indicate imputed not. Results correlational specification regresses isolation. step, jointly subsequently relationships. signals chances grow obtaining favourable (including domains). results, correlational, informative evidence holding constant. reflect unobservable characteristics. issue Sections 5.2 5.3. 13 1. portrayed graphically Figure figure (Model 1), finally additionally specifications, Tables A4 A6. detail. 5.1 Main estimation 5.1.1 Reading shown left quadrant A3. Not surprisingly, strongest domains. increase reduction 0.29 deviation. remain statistically significant included, coef- ficient suggests 0.10 per increase. With included. last rows marginally reduces coefficients. appear mechanism measure. 5.1.2 Educational mem- 14 ber, (different of) categorical, distinguishing signifi- cantly specifications controls. especially, associated decrease 0.7 corresponds 0.25 deviation). Including severely estimates. longer full (mainly income). coefficient -0.22. drive further, previous exist high Achievement appears (both domains) noncognitive cutoff degree levels. attributed alternatives, comparable. Among GCSE A-C connections Dummy end 15 distribution weaker 5.1.3 Adult during right A5. smaller here. small margin. (‘separate’) ranking percentiles. added, decreases 0.9 percentile. compared initial estimate. specification, Controlling coefficients, possession Part operates skills, mimics 42) incomes, highly consistent 5.1.4 Health 42, Interestingly, dominate Social adulthood. 0.136 five-point scale (and 1). initially association hardly (observed) change added (once occur taken. adulthood, questionnaires predate physical problems prominent.16 selective controlling class). Similar mediate 5.1.5 Non-linearity assumed now linear. worthwhile explore whether, extreme impact, need reach before effect. nonlinearities subsection, estimating polynomials domains.17 non-linearity. apparent Comparing inhibits non-linear tendencies. quadratic positive, indicating negative diminishing. deprived. possibly skewed left. implies distribution. fits hous- 16See, Kessler (2007). 17These request. 17 ing interesting (especially) severe non-linearities certainly involve sign reversal non-linearity surprising, linear begin with. case, attenuated fit. Finally, interaction complementarity neighbourhoods, vice versa. 5.1.6 assessed noteworthy summarized A7. Coefficients (the size 0.1 well-being), possessional, (immaterial) Estimates satisfaction. Furthermore, acts relationship, rooms house highlights persistence proper Body Mass Index (BMI). BMI allows 42 positively negatively 18 appliances microwaves) less diet. weaker, lag 10. adolescence. Value-added Because points lagged model. earlier achieved growth exercise one. lags half, positive lags. math low, negative. contrasting remarkable. electronics, bedroom 19 opposite scores.18 measures, line revealed score, extent. control, variable, shaped exhibit adolescence adulthood.19 5.3 unobserved investments, identified biased. influence (GSA) extended Harada continuous unobservables required 18We speculate underlying reasons result. One watching complementing subjects (Borzekowski Robinson, 2005; Sharif ways development. 19See, Cunha (2010); (2011). 20 insignificant.20 plausibility parameters partial R2 ’s parents, ethnicity structure, observable. needed render insignificant combined, plausibly On hand, if away , selection. plots plausible cases: requires estimate, straightforward. so (unobservable) weak best. GSA produces emphasized variance explained plotted graph conservative, addition 20One value. 21 implausible, even though (very slightly) X.21 adding classroom peer 0.01, 0.017. case. representative X curve scores, self-esteem. area 5.4 Explanatory valuable joint marginal reveals uniquely explains (extensive) student 21Additionally, statistical significance (at 10% level), 0. majority cases conclusion condition, lines above 22 before, minor, means (especially self-esteem) compare ‘gross’ ‘net’ latter. A1 look reversed). naturally shares limited. figures con- firms adverse 5.5 better essence argued bands subject error. Keeping mind, portray restrict overall Correlations slightly perform (an imperfect eliciting at-risk 23 modest third fourth column split pattern indicator; subset predictive math, second except again are, average, achievement. significant. dominance remarkable (strictly) (likely attenuated) ‘traditional’ error.22 weakest characteristics). worth noting here, lose multidimensional advantages. unique seven defined, 22One focused bottom broad, distinguishes (its close normal, long tails side). similar, topcoded. 24 siderably does, 5.6 Differences constructed indicators, too robustly age-effects Robustness model, assumptions relax assumptions. 6.1 Bad problem away. styles, divorces, 25 downwardly 10; 5; birth. ‘bad controls’ specifications. Deprivation incorporating surprising Nonetheless, panel B contribute downward influenced Conversely, excludes controlled for, test, 6.2 Affordability count it. executed belong (these 26 domain). 5. unconditional Only looking simply owning connection (possessional) 6.3 Different specified restriction cannot limit, society. 50% 25% 15%. tighter remove 20% 40% respectively. fall changes. 25-50% 15-50% strongly. changes restrictive almost Sensitivity thresholds gradually three models 15% approach. 27 6.4 Endogenous items? present misleading estimates, process way around. visits museum, library plays musical instrument, partially interest (partially) parent-child activities, member club. estimation. Especially expect interest, however, lies relatively. A9 proportional little. zero. significant, positive. achievement.23 ex- 23The restricted subdomains: neighbours financial constraints. responsible scores. 28 ercise should, interpret care. determine truly resulting (also) personality independently state household. representing (perceived) environment support surrounding 6.5 Attrition heterogeneity members disappear altogether Further attrition non-random. Those differ Most prominently, male (58.4% vs. 49.4%), non-native (16.8% 8.6%) (12.2% 5.6%). fully population period. external validity may selective. turns gender. boys. out-of-wedlock None large, similarly moderate loss representativeness 29 described imputed. employing conventional identifying attenuation sensitivity, interactions corresponding (thereby allowing intercept slope respective variable), A8 approaches: applied estimation, sum binary latter, remarkably small, (commonly used) relating sizes Judging fit, 30 6.7 Financial hardship broad. incorporated lacking namely ability face unexpected expenses arrears bills. directly, ask were troubled past year. insignifi- cant Conditional percentage 0.100 Conclusion experienced reveal Plausible adds (mental) analyzed ‘deprivation’, form (a capital, (conditional) 31 diminish environment. Previous huge children. Our households, isolated, contributor literature, impacts. disconnect ‘material state’ with, causally to, supportive McLanahan Bianchi (2006) educated fathers mothers style contributors Research investments low-income improved changing guidance beliefs, relief (Kautz 2014). ‘immaterial (imperfectly measured) nature. At invalidate alto- 32 gether. Basic (although groups things, improves identifier limitations paper. automatic taking life-time perspective changed inevitable linking achievements. causality likely) exploitation elements help segments Future precisely links. greatest challenge exactly why obtain widely References Akee, R. K., W. E. Copeland, G. Keeler, A. Angold, J. Costello Parents incomes children’s outcomes: quasi-experiment. American economic journal. Applied economics 2(1), 86–115. Almlund, M., L. Duckworth, T. D. Kautz Personality psychology economics. Handbook Education, Volume 4, pp. 1–181. Amsterdam: Elsevier. 33 Beckett, C., B. Maughan, M. Rutter, Castle, Colvert, C. Groothues, Kreppner, S. Stevens, O’Connor, Sonuga-Barke Do persist adolescence? Findings English Romanian adoptees study. Child Development 77(3), 696–711. Bianchi, P. Milkie Changing Rhythms Of Family Life. New York, NY: Russell Sage Foundation. Bjorklund, A., Lindahl, Plug (2006, August). origins asso- ciations: Lessons swedish Quarterly Journal 121(3), 999–1028. Blanden, attainment: review Britain. Oxford Review Economic Policy 20(2), 245–263. Borzekowski, N. Robinson (2005). remote, mouse, no. pencil: media academic grade students. Archives Pediatrics & Adolescent Medicine 159(7), 607–613. Bradley, H. F. (2002). Socioeconomic Annual 53(1), 371–399. Brooks-Gunn, (1997). 55–71. Cappellari, Summarizing ISER Working Paper Series, 2006-40. Carneiro, (2003). Human capital policy. Krueger, Friedman (Eds.), Inequality America: What Role Capital Policies?, 77–239. Cambridge, MA: MIT Press. Chevalier, Harmon, V. O’Sullivan, I. Walker (2013). schooling IZA Labor 1–22. 34 Corak, Income inequality, equality opportunity, mobility. Perspectives, 79–102. Cunha, F., Schennach Estimating technology skill formation. Econometrica 78(3), 883–931. Dearden, L., Emmerson, Frayne, Meghir (2009). dropout rates. Resources 44(4), 827–857. McGranahan, Sianesi Credit Constraints Choices: NCDS BCS70. CEE DP 48. ERIC. Ermisch, Origins immobility inequality: National Institute 205(1), 62–71. Commission 2020: smart, sustainable inclusive growth. Filmer, (1999). 35 Population 25(1), 85–120. Frijters, P., Haisken-DeNew, Shields health: german reunification. 24(5), 997–1017. Fusco, A.-C. Guio, Marlier Technical report, CEPS/INSTEAD. Harada, (2012). Generalized NYU working (2000, March). Policies foster human capital. 54(1), 3–56. (2008, July). Schools, synapses. Inquiry 46(3), 289–324. Imbens, (2003, May). exogeneity evaluation. 93(2), 126–132. Kautz, T., Diris, Ter Weel, Borghans (2014). Fostering skills: Improving promote lifetime Bureau Research. Kessler, Amminger, Aguilar-Gaxiola, Alonso, Lee, Ustun Age onset disorders: literature. Current opinion psychiatry 20(4), 359–364. Løken, K. education: norwegian experiment. Labour 17(1), 118–129. McKnight, (2015). Downward mobility, hoarding glass floor. report. McLanahan, Diverging destinies: faring demographic transition. Demography 41(4), 607–627. non-monetary exclusion: Europe? Analysis Management 29(2), 305–325. new tool monitoring (child) poverty: cumulative Indicators 5(2), 335–355. Society Glance: 2006 Edition. Organisation Co-operation Development. Together: Why Less Benefits All. Paris, France: Publishing. Perry, Zealand, 101–127. Sacerdote, Nature nurture learned twins adoptees? Economics. NorthHolland. 36 Sharif, I., Wills, Sargent Effect visual performance: prospective 46(1), 52–61. Tomlinson, Walker, Williams concept, 1991 2003. 37(04), 597–620. Whelan, Layte, Understanding dynamic comparative Sociological Review, 287– 302. Vulnerability perspectives europe: latent class Societies 7(3), 423–450. Woßmann, equal opportunities? US. CESifo 1162. 37 1: Possess House Neighbour Educ (M) (I) −.3 −.2 −.1 .1 −.6 −.4 Educ. att. −3 −2 −1 −.05 .05 Raw C C+M Notes: reading. portrays (‘Raw’), (‘C’) (‘C+M’). ‘C+M’ horizontal bars 95% confidence intervals. equally spaced categories. averaged expressed percentile range. 2: .2 .4 .6 Partial .8 EV −−> P HE .3 .5 HO (partial (‘EV’; deprivation) level. (D) (X0 Figures (C (I)) (all row), (P) (HE) (HO) row). 39 3: Math Self−esteem Mental Life satisfaction Controls Domains (vector 1) regressing 40 Possession Housing Neigh Panel A: VA Reading16 0.035* 0.003 0.001 -0.030*** -0.050*** -0.107*** -0.001 (0.021) (0.018) (0.010) (0.011) (0.012) LC 0.051*** 0.016 -0.006 -0.018* -0.029** -0.054*** 0.002 (0.019) (0.016) (0.009) Reading10 -0.051*** -0.020** -0.026** -0.138*** -0.008 (0.014) -0.031** -0.019* -0.021** -0.110*** -0.005 (0.013) Math16 -0.066*** -0.002 -0.014 -0.059*** -0.049*** -0.093*** -0.024 (0.025) (0.020) (0.015) -0.040* 0.019 -0.010 -0.038*** -0.027** -0.030* -0.022* (0.017) Rutter16 -0.007 -0.056*** -0.028*** -0.032*** -0.046*** -0.065*** -0.047*** -0.025*** -0.035*** -0.024** -0.028** -0.044*** Rutter10 -0.036** -0.019** 0.007 -0.070*** -0.020 0.009 -0.009 -0.048*** (0.008) Locus16 -0.042** -0.016 -0.019 -0.027 -0.041*** (0.023) 0.014 -0.038** -0.015 -0.023 -0.037*** Self-esteem16 -0.030 0.026 -0.033*** -0.047** -0.161*** 0.029 -0.046** -0.157*** B: Reading5 -0.077*** -0.034*** 0.004 -0.004 -0.111*** -0.011 Math10 -0.057*** -0.022** -0.018** -0.042*** -0.121*** Rutter5 0.041*** 0.053*** 0.026*** 0.006 0.015 0.027*** Locus10 -0.023** -0.003 -0.045*** Self-esteem10 -0.027*** -0.025** table ‘LC’ (taken outcomes) available. regressions. ‘Rutter’ scale. ‘Locus’ internal 41 No Y (all) 0.801*** -0.818*** -0.434*** -0.594*** 0.179*** -0.295*** -0.127*** -0.249*** (0.035) (0.039) (0.046) (0.043) (0.042) (0.038) 7.29*** -7.26*** -5.20*** -3.30*** 3.66*** -2.51*** -0.999** -2.34*** (0.396) (0.399) (0.436) (0.440) (0.484) (0.450) (0.439) (0.382) 0.154*** -0.179*** -0.113*** 0.054*** -0.087*** -0.055*** 0.242*** -0.218*** -0.172*** 0.042* -0.026 0.010 (0.024) (0.022) 0.301*** -0.335*** -0.176*** -0.244*** 0.038** -0.131*** -0.124*** 0.279*** -0.307*** -0.213*** 0.079*** -0.149*** -0.100*** -0.082*** (0.029) (0.027) (0.026) 0.130*** -0.197*** -0.123*** -0.119*** 0.026* -0.072*** -0.079*** 0.129*** -0.258*** -0.114*** -0.231*** 0.041 -0.219*** -0.184*** 0.898*** -1.25*** -0.468*** -1.27*** 0.291 -0.790*** -0.240 -0.937*** (0.138) (0.157) (0.162) (0.161) (0.193) (0.194) (0.184) (0.169) (Y) Effects separately. (all), (I). estimations Possession16 Possession10 Possession5 Housing16 Housing10 Housing5 -0.036 -0.035 -0.032 -0.033 0.028 (0.036) (0.031) -0.192 0.342 -0.951*** -0.872*** -0.783*** 0.158 (0.311) (0.395) (0.329) (0.245) (0.246) (0.253) 0.012 0.039* -0.034* 0.025* Age-specific 4: Exclusion -0.209*** -0.418*** -5.13*** (0.312) lim -2.09*** (0.288) -0.028 -0.137*** -1.82*** (0.028) (0.317) -0.288*** -0.730*** -5.57*** -0.136*** (0.301) -0.064*** (0.297) -0.208*** -2.43*** (0.306) -0.264*** -0.513*** -4.03*** -0.112*** (0.310) -0.242*** -0.463*** -3.76*** -0.098*** (0.019 (0.279) -0.043* -0.062** -1.23*** (0.305) -0.203*** -2.65*** (0.030) (0.341) -0.205*** -2.50*** (taking together) 43 5: Lack (A) (B) -0.426*** 0.027 -0.423*** -0.0024 0.025 -3.49*** -1.31*** 0.022 -3.48*** -0.729*** 0.018 (0.286) (0.307) (0.275) 0.0093 -0.089*** -0.026*** 0.017 (0.0098) (0.0097) Add -0.210*** 0.054** 0.104 -0.196*** 0.075*** 0.101 -1.77*** -0.574** 0.072 -1.63*** -0.118 0.069 (0.303) (0.283) (0.273) 0.030 -0.017* 0.0065 -0.040** 0.00011 (0.0095) -0.059** 0.235 0.034 -0.434 -0.740*** 0.225 -0.405 -0.188 0.224 (0.295) (0.265) (0.298) (0.251) -0.0027 -0.013 0.054 -0.0058 0.053 0.0055 0.248 0.0098 0.00018 (0.0091) compares afforded (B). 44 6: maximally allowed (main) -0.083*** -0.025 -0.031 -0.088*** -0.061** 0.231 -0.790** -0.922*** -0.537** -0.630*** -0.310 -1.718*** -1.010*** 0.324 (0.333) (0.272) (0.228) (0.232) (0.249) (0.271) -0.053*** 0.047 0.013 -0.103*** 0.237 -0.067** -0.085*** -0.060*** -0.226*** -0.039* 0.229 -0.416 -0.826*** 0.507** -0.433* -0.303 -1.674*** -0.615** 0.321 (0.313) (0.229) (0.233) (0.239) (0.262) (0.250) -0.012 0.028*** -0.036*** 0.008 0.011 0.236 -0.038 -0.181*** 0.226 -0.771** -0.648** 0.120 -0.399* -0.762*** -1.178*** -0.257 0.319 (0.319) (0.269) (0.237) (0.231) (0.260) (0.256) -0.016* -0.031*** 0.045 0.005 -0.099*** 0.232 top 50%, 25%, regressions 45 A1: variance: (reversed) measures. 46 Items • Possessions: (does own:) refrigerator, washing machine, dryer, TV, car, phone, video recorder, camera, stereo, radio, PC, sewing vacuum cleaner, microwave 16); freezer, holiday 10); phone 5); child: cassette player, bicycle 16) Health: eats meat fish times week, breakfast, lunch, iron, vitamins, milk, fibre, carbons, sugar intake breakfast 10) Neighbourhood: noisy, graffiti neighbourhood, youth loitering streets, drunks rubbish street, victim crime, beak-in, unsafe night (age 5) Housing: bathroom, indoor toilet, hot water, garden, kitchen, bed, difficulties heating house, moisture problems, untidy furniture state, (material): played books home, studying, newspapers, calculator constructional toys (immaterial): read little, library, Social: neighbours, participate activity (excursions, charities, concerts), talk rarely participates misses money, club organization 16), friends, 47 A2: Control Birth controls: abnormalities, hospital born, head circumference, mother married father Household income: eligibility free lunch class: employment: works (averaged hours worked 5), work experience structure: people older siblings, younger siblings Parenting style: attitude toward gender independence authoritarian world smoking (ever), pregnancy, heavy pregnancy 48 A3: Separate -0.128*** -0.169*** -0.289*** – -0.044** -0.096*** -0.193*** 0.115 0.000003 -0.039*** -0.017 0.181 0.000 -0.020* -0.167*** -0.021* 0.132 -0.081*** -0.170*** 0.144 Parent empl -0.041** -0.029*** -0.091*** -0.198*** 0.119 structure -0.031* -0.188*** 0.141 -0.090*** -0.165*** 0.133 Non-cog (NC) 0.021* 0.249 (WC) row regressed separately, jointly. isolation). Rows (WC). Row See contained A2 49 A4: -0.560*** -0.449*** -0.260*** -0.451*** -0.707*** -0.433*** -0.271*** -0.140*** -0.162*** -0.486*** 0.106 -0.261*** -0.130*** -0.171*** -0.456*** 0.118 0.198 -0.074** -0.056** -0.057** -0.135*** -0.390*** -0.101*** 0.147 -0.153*** -0.055** -0.126*** -0.399*** -0.129*** 0.159 -0.239*** -0.147*** -0.462*** -0.104*** -0.092*** -0.148*** -0.160*** -0.425*** 0.142 -0.187*** -0.071*** -0.132*** -0.394*** 0.137 -0.058*** -0.192*** -0.080*** 0.243 -0.047 -0.073*** -0.045* 0.301 -0.078*** -0.141*** -0.429*** 0.128 -0.018 -0.194*** 0.242 -0.037* 0.303 50 A5: -5.17*** -4.12*** -2.81*** -3.04*** -4.17*** -4.82*** -2.89*** (0.299) (0.282) (0.254) (0.266) (0.278) (0.268) -2.630*** -1.700*** -1.540*** -2.392*** -2.158*** -2.576*** -0.454 0.071 (0.353) (0.255) (0.264) (0.308) -2.260*** -1.583*** -0.971*** -0.956*** -0.851*** -3.238*** -1.399*** 0.276 (0.321) (0.284) (0.281) -1.651*** -1.394*** -1.140*** -2.097*** -1.611*** -1.383*** -0.153 (0.354) (0.309) (0.261) -0.901** -1.055*** -1.239*** -2.231*** -1.889*** -1.798*** -0.088 0.099 (0.371) (0.314) (0.276) -2.148*** -1.552*** -1.300*** -2.208*** -1.929*** -2.071*** -0.462 0.090 (0.358) (0.277) (0.285) -2.381*** -1.592*** -1.537*** -2.394*** -2.178*** -2.552*** -0.419 (0.364) (0.315) (0.287) -2.562*** -1.570*** -1.527*** -2.374*** -2.070*** -2.527*** -0.368 (0.360) -2.052*** -1.469*** -1.376*** -2.255*** -2.016*** -1.964*** -0.326 0.084 (0.357) -1.684*** -1.085*** -1.314*** -1.838*** -1.425*** -0.544* -0.116 0.129 (0.348) (0.304) -0.639* -0.726*** -0.516** -0.270 0.077 -0.600** -0.976*** 0.363 (0.326) (0.221) (0.242) (0.243) -2.572*** -1.612*** -1.371*** -2.182*** -2.046*** -2.402*** 0.052 0.083 (0.352) (0.292) -0.792** -0.857*** -0.401* -0.476** -0.222 -1.526*** -0.549** 0.332 (0.332) (0.270) -0.650** -0.710*** -0.204 0.113 -0.563** -0.699*** 0.366 description 51 A6: -0.108*** -0.125*** (0.0099) (0.0092) (0.0093) -0.040*** 0.031 0.038 -0.030** 0.036 -0.068*** 0.037 -0.069*** 0.035 0.044 0.048 0.059 0.061 Subjective 52 A7: Arrests34 0.030*** (0.007) (0.006) health42 -0.300** -0.117 -0.066 -0.262*** -0.105 -0.709*** (0.147) (0.121) (0.095) (0.098) (0.101) (0.122) (0.103) satis34 -0.043** -0.150*** (0.120) Finances42 0.024* 0.033*** 0.024** BMI42 -0.276*** 0.190*** 0.345*** -0.057 0.057 (0.091) (0.073) (0.059) (0.063) (0.064) (0.078) (0.065) -0.278** -0.034 0.221*** 0.239*** 0.088 -0.042 0.240 (0.115) (0.099) (0.060) (0.076) (0.071) BMI16 -0.053 0.197*** 0.040 -0.046 (0.048) (0.032) (0.053) 0.060* 0.060 0.067* 0.275 (0.047) BMI10 0.020 0.151*** 0.046 BMI42) ‘BMI’ member. ‘Arrests34’ arrests 34.‘Mental health42’ score Warwick Edinburgh well-being scale, ‘Finances42’ situation 53 A8: Comparison (educational attainment) Factor PW -0.332*** -0.191*** -0.357*** -0.122*** 0.093 Sum -0.292*** -0.158*** -0.037 -0.517*** -0.118*** 0.086 Binary -0.633*** 0.129** -0.527*** -0.982*** -0.199*** (0.057) (0.056) (0.058) -0.095*** -0.049** -0.224*** -0.099 -0.138** -0.476*** -0.130** -0.065 0.228 (0.061) (0.055) (0.054) (baseline) ‘Factor’ ‘PW’ ‘Sum’ sums applies. ‘Binary’ 0/1 75th labeled deprived). 54 A9: Excluding Exclude -0.062*** -0.166*** 0.227 -1.000*** -1.028*** -0.573** -0.714*** -0.286 -1.166*** 0.320 (0.334) (0.230) (0.248) (0.234) -0.021 0.023 partly member) excluded. exclusively 55 Copyright © @ author(s). Discussion papers draft form. distributed purposes comment only. reproduced permission copyright holder. Copies author.