This study models poverty changes across the United States between 1990 and 2000 as
a function of spatial poverty variables (variables that attempt to capture the spatial
effects of poverty and poverty changes), social capital, local employment growth, and
demographic controls. Newly available decennial census data at the tract level is used
to capture the benefits of a more homogeneous geographic aggregation. The spatial
variables account for the effects of own poverty, poverty pockets, and dynamic spatial
spillovers on poverty changes. Local economic and social context variables include
measures for social capital and employment growth. Spatial econometric model
formulations are compared with OLS, and results indicate that spillovers (a spatial lag
variable) as well as a spatial error correction are significant. Overall, being in a
poverty pocket (being surrounded by high poverty tracts) had a detrimental effect on
tract outcomes, while employment growth and social capital had positive ones.
Interaction terms are used in a second model to assess the effects of initial own
poverty level on poverty change determinants. Significant spatial lag and spatial error
correction parameters again indicate that the fates of neighboring tracts are tied
together. Results from the interaction model show that there are significant differences
in poverty determinants based on tract poverty rate at the beginning of the period.
Having higher levels of minorities or poverty pocket location had a negative effect on
poverty reduction in low poverty tracts, while college education rates had a positive
effect. Poverty reductions in high poverty tracts were speeded by employment growth
and social capital. The results suggest that anti-poverty policies, to be most effective,
need to be sensitive to the effects of geography, and that employment growth is
effective in reducing poverty, particularly in high poverty areas.
File scanned at 300 ppi (Monochrome, 24-bit Color) using Capture Perfect 3.0.82 on a Canon DR-9080C in PDF format. CVista PdfCompressor 4.0 was used for pdf compression and textual OCR.
description.provenance : Made available in DSpace on 2011-02-17T17:58:13Z (GMT). No. of bitstreams: 1
CrandallMindyS2005.pdf: 1260339 bytes, checksum: 1e75bb5fe6afb61f790f97dc8b3af3db (MD5)
Previous issue date: 2004-06-17
description.provenance : Submitted by Sergio Trujillo (jstscanner@gmail.com) on 2011-02-15T22:51:26Z
No. of bitstreams: 1
CrandallMindyS2005.pdf: 1260339 bytes, checksum: 1e75bb5fe6afb61f790f97dc8b3af3db (MD5)
description.provenance : Approved for entry into archive by Patricia Black(patricia.black@oregonstate.edu) on 2011-02-17T17:58:12Z (GMT) No. of bitstreams: 1
CrandallMindyS2005.pdf: 1260339 bytes, checksum: 1e75bb5fe6afb61f790f97dc8b3af3db (MD5)
description.provenance : Approved for entry into archive by Patricia Black(patricia.black@oregonstate.edu) on 2011-02-17T16:29:14Z (GMT) No. of bitstreams: 1
CrandallMindyS2005.pdf: 1260339 bytes, checksum: 1e75bb5fe6afb61f790f97dc8b3af3db (MD5)