Lori Curtis and Kate - PowerPoint Presentation

Lori Curtis and Kate RybczynskiDepartment of EconomicsUniversity of WaterlooOctober 24, 2012Acknowledge SSHrC for funding. No Exit: Women in Poverty

MotivationWomen face higher risk of long term poverty.(Finnie & Sweetman 2003; Lochhead & Scott 2000; Burstein 2005)Women comprise ~61% of the long term poor Few studies on determinants of poverty duration and exits in Canada. (Finnie & Sweetman 2004; Finnie 2000; Antolin , Dang & Oxley 1999; Burstein 2005; Lochead & Scott 2000) Negative Duration Dependence Family Composition Employment Status Age Education (Welfare studies & US poverty studies)

Data & Methodology Canadian Survey of Labour and Income Dynamics (SLID) 1994-2006 (Panels 1-4) target population is all individuals in Canada excluding persons living on Indian reserves, institutionalized individuals and some northern communities (less than 3% of the population). The sampling frame is the Labour Force Survey. Interviewed yearly between January and March regarding labour market experiences, income, education, family relationships and other demographics Data were accessed in the Southwestern Regional Data Centre at the University of Waterloo which is part of the Canadian Data Research Data Network. Although the data were accessed through Statistics Canada, the opinions sited within do not reflect Statistics Canada’s policies or opinions.

Data & Methodology Poverty spell: continuous period household income falls below the after-tax LICO. Exclude spells where individual under 25, over 60 or a studentunder 25 or students may be poor given their current income but long-term outlook is very different than others living in poverty over 60 due to retirement issues and the possible receipt of government provisions (OAS and/or GIS). Those with missing information also excluded. Sample of 2953 poverty spells for 2764 women

Duration of Poverty SpellDuration is end date less start date in years Start dates unknown if spell starts before first year of panel. End dates unknown if spell continues after last year of panel. Flagged as left-censored and right-censored, respectively (Finnie and Sweetman, 2003) Right-censored spells retained Left-censored spells are omitted from the sample No characteristics of the individual on entering poverty

MethodologyDescriptiveProportional hazard, discreteModels with and without gamma distrib. heterogeneity Prentice-Gloeckler 1978; Jenkins 2008 Analysis by Exit type (competing risks) Report exponentiated coefficients = relative risks for all models

MethodologyConstruct our competing risks analysis by generating a categorical variable equal to 0 if the individual does not exit in time t, 1 if they exit to less than 1.1 times the LICO, 2 if they exit to an inclusive range of 1.1 and 2 times the LICO,3 if they exit to more than twice the LICO. Allison (1982) Specification 1 control variables measured prior to entering povertySpecification 2 adds some control variables measuring changes within the poverty spell (e.g., become married, now have children, begin full-time work, etc)

Table 1 Spell Characteristics Sample a Censored b Average Spell Duration 1.865 (1.131) 1.405 (0.717) Percent with Multiple Spells 18.150 20.750 Percent Exiting Poverty 69.180 100 Percent Right Censored 30.820 0 Number of Spells 2953 2043 Note: standard errors in brackets. a includes cells that are right censored but excludes spells that are left-censored b excludes cells that are right and left censored

Distribution of Ratio of Family Income to LICO on Exiting Poverty Spell

Table 2 Spell Characteristics by Family Income Relative to LICO on Spell Exit Inc < 1.1*LICO 1.1* LICO ≤ Inc≤2*LICO Inc > 2.0* LICO Average Spell Duration 1.542 (0.829) 1.415 (0.717) 1.252 (0.566) Percent Multiple Spells 27.080 21.360 13.320 Percent Exiting to 18.8 60.250 20.950 Note: standard errors in brackets.

Descriptive Stats show1/3 of poverty spells do not end in the study periodaverage duration of  poverty spells is almost 2 yrs. 1/5th of  the sample has multiple poverty spells. close to 20% exit to <110% of the LICO, 60 percent exit to ‘near poverty’ (1.1 to 2*LICO)only 20%  exit to higher-income levels (>2* LICO). ¼ exits < 2*LICO by women with multiple spells.

Duration analyses ‘determinants’ of exiting poverty. Factors (year prior to poverty spell) that increase likelihood of exitLabour attachment year prior to poverty spellhaving higher education Decrease likelihood of exit participating in social assistance being an immigrant having younger children being unattached (compared to couples with no children) years in poverty increase (negative duration dependence)

Adding Characteristics that change within a spell results in substantively similar estimates All change variables decrease likelihood of exit as possibly decrease incomeleaving a marriage having children enter or leave the household becoming disabled are associated Multiple spells increase likelihood (have to get out to get back in)

Competing Risks Framework (compared to not exiting)SA prior to entering poverty less likely to exit to 1.1*LICOFull year employment and having a high school diploma (compared to < HS) are associated with moving out of poverty to income that is between 1.1 and 2*LICO.Higher levels of education, employment and age are positively associated with exiting to >2*LICO being an immigrant, unattached, on social assistance, or having young and more children are negatively associated with leaving poverty to>2*LICO

Table 5 Mean Spell Duration pre and post Intro of NCB Spells starting before 1998 Spells Starting in or after 1998 a All Females 1.992 (1.254) 1.7803 (1.036) Lone Parents 2.200 (1.272) 1.730 (1.006) Couples with Children 1.862 (1.170) 1.699 (0.995) Other Family with Children 1.964 (1.333) 1.981 (1.019) Unattached2.226(1.345)1.976(1.078)Other Family - no Children 2.130(1.352)1.794(1.085)

Possible Policy ConclusionsOur data seem to indicate that a not so small portion of women who enter poverty are ‘trapped’ there. increasing education and employment opportunities for the poor. temporary or longer-term guaranteed living income for those with disabilities, larger families and long-term social assistance recipients who are not able to find employment or women with multiple spells

Possible Policy Conclusionsthe competing risks framework demonstrates that exiting poverty is not the same experience for all women. Studies examining poverty and poverty duration should differentiate between those who exit to near the poverty line or to far above the poverty line – clearly, these are very different experiences for women and their families and are strong indicators as to whether or not a women (and her family) will return to poverty.