Neo-liberalism at the Subnational Scale: Local Governments
Growth and Redistribution Policy in an Era of Decentralization
POVERTY AND PLACE ACROSS THE UNITED STATES:DO COUNTY GOVERNMENTS
MATTER TO THE DISTRIBUTION OF ECONOMIC DISPARITIES? Presented by:
Mark D. [email protected];
http://aede.osu.edu/programs/Swank/
Coauthors: Linda Lobao, Wilner Jeanty, & David Kraybill
Keynote: Poverty and Regional Analysis Workshop50th Annual
Meetings of the Southern Regional Science Association
Pre-ConferenceJ. Norman Efferson Series LectureLouisiana State
University February 3, 2011This research received partial support
from NIH award R21-HD47943 to the Initiative in Population Research
at The Ohio State University.Partridge--LSU1Purpose of the
StudyContribute to the poverty-and-place literatureat the
sub-national scale. First study to question the impacts of counties
as local governments:Are county governments a missing link in this
literature?Do county governments influence local populations
economic well-being?Most novel part in my opinion is whether it is
good policy that matters or government capacity that matters. Focus
on job growth, poverty rates, and household income in the 2001-2007
period.
Partridge--LSU2Outline of the PresentationI. The
poverty-and-place literature and countiesII. More interest in local
govt role in reducing povertyIII. Conceptual debate about the
effects of local govts on poverty and other forms of well-beingIV.
Empirical analysis: Do county govts influence local well-being?
Focus on countys institutional capacity and policy initiatives
directed to workforce developmentNational analysis of county govts
using secondary data and unique primary survey dataV.
ConclusionsPartridge--LSU3Spatial Clustering of PovertyPersistent
patterns that is worse near the
DeltaPartridge--LSU4Partridge--LSU5
Source:
http://www.census.gov/did/www/saipe/data/statecounty/maps/2009.html
downloaded Jan 20, 2011.5
1969197919891999
Family Poverty (terciles) Partridge--LSU6Family Poverty
Clustering (Local Morans I)
Partridge--LSU7I. Poverty across U.S. Localities:
LiteratureThematic concern with factors explaining sub-national
economic disparity (e.g. poverty rates, income levels, or income
inequality)Common independent variables (see paper for cites):
economic structure (quality & quantity of local employment)
demographic attributes (e.g. age, race/ethnicity, family structure
reflecting vulnerability to poverty) geographic (agglomeration,
regional context) Methodological protocols:counties most commonly
used unit-of-analysisquantitative studies using various
regression-type modelsPartridge--LSU8Poverty across U.S.
Localities: LiteratureCommon limitations in this literature:Weber
et al. (2005); Partridge and Rickman (2006) Studies vary in degree
to which spatial processes incorporatedspatial econometrics, GWR,
etc. Endogeneity and self-sorting are not usually considered.
Relatively little attention to govt, even less to local govt
Partridge--LSU9Partridge--LSU10Poverty across U.S. Localities:
LiteratureCounties the empirical sample of choice:Isserman et al.
2009; Peters 2009 mostly fixed geographic boundaries considerable
secondary data spatial scale(incorporates residence and work, most
people live and work in the same county) counties include both slow
& fast growing (urban/rural) places (Desmet and Fafchamps
2005)
Local Poverty: The Body of LiteratureCounties are also important
units of govt:Lobao and Kraybill, 2005 cover more residents than
other local governments, such as municipalitiesnow fastest growing
U.S. general purpose governmentprovide important human services and
administer welfare (the TANF program) for many Americans
Partridge--LSU11Poverty across U.S. Localities: LiteratureWhy
are counties overlooked as units of govt? Conceptual gaps in the
poverty-and-place literature evolved with a focus on economic and
demographic determinants Federal government receives most
attentionWar on Poverty most vivid example Americans see the state
as the federal stateSheingate (2010); (Peterson 1981) Lack of
county govt data (Weber et al., 2005)
Partridge--LSU12II. More Interest in Local Govt Role in Reducing
Poverty Policy interest in local govts from both right and left
sides of political spectrumThe role for local govts to help reduce
poverty has gained traction with greater interest in social
capitalDecentralization has increased the functional
responsibilities of local govts. Rodriguez-Pose and Gil (2005) New
policy-experiments undertaken by local govts Place-based poverty
policy increasingly recognized by some as an important adjunct to
people-based policy. Partridge and Rickman (2005, 2006)
Partridge--LSU13III. Local Govt and Economic Well-Being:
Differing ViewsThe Case for the Beneficial Effects of Local
GovtGovt has a role in society beyond providing pure public goods,
e.g., greater economic growth and reducing poverty. Bartik (1991);
Lobao and Hooks, (2003); Volscho and Fullerton (2005)1. The
Institutional Capacity of Local GovernmentGovernment provides two
overlapping functions: --an economic growth function (job
creation)-- a social equity function: strengthens the local social
safety net, protects residents from destabilizing market
forcesWhere govt is institutionally stronger (i.e., with greater
bureaucratic and fiscal capacity to operate effectively), its
beneficial influence is greaterpoverty should be lower and
household income higher. Lobao and Hooks 2003
Partridge--LSU14III. Local Govt and Economic Well-Being:
Differing ViewsThe Case for the Beneficial Effects of Local Govt 2.
Specific Policies and Programs of Local GovtTiebouts (1956);
Rodriguez-Pose and Kroijer (2009) Economic development programs
aimed to local business and public services directed to the local
labor forceSuccessful economic development programs should create
jobs, in turn, raising family incomes, and reducing poverty
Programs directed to community workforce: can build human capital,
promote family well-being, increase community cohesion, and reduce
poverty (e.g., transportation and childcare) (Blank
2005).Partridge--LSU15III. Local Govt and Economic Well-Being:
Differing ViewsThe Case for Null or Negative Effects of Local Govt
Researchers build their arguments from one of two vantage points,
grounded in policy-schools from the right and the left. A. The
limited-govt view sees market and state as intrinsically
competitive, emphasizes govt failures, see Okun, (1975); Buchanan,
(1986) As local govts expand institutionally beyond providing true
public goods and to undertaking social equity functions, they
adversely affect local culture and labor supply and can turn
localities into welfare magnets (e.g. Murray 1984). Programs
directed to low-income people are particularly misdirected in
catering to underperforming/undeserving populations, which over the
long term undercuts growth. Partridge--LSU16III. Local Govt and
Economic Well-Being: Differing ViewsThe Case for Null or Negative
Effects of Local Govt B. Critical political economy literature view
of local govt Recognizes govt as an important institution for
ensuring well-being and alleviating poverty but stresses the
inadequacies of local govts. Critique leveled against devolution of
federal welfare (TANF)/social programs to program to states and
counties. (Schram 1999) Also critique leveled against local
economic development programs directed to businesslocal elites
rather than low-income people capture any gains (Storper and
Manville, 1986; Urban Studies)
Partridge--LSU17Partridge--LSU18III. Local Govt and Economic
Well-Being: Differing ViewsIn sum, we test two opposing hypotheses
for county govts: Benefits of county governmentWhere county govt is
larger and institutionally stronger (i.e., with greater
bureaucratic & fiscal capacity), its beneficial influence is
greatergreater job growth & income & lower poverty
rates.County-govt programs directed to business and to the
community workforce should be related to greater job growth, lower
poverty, and higher household incomes. Related Hypothesis: Is it
Policy or Govt Capacity that matters?
Limitations of county governmentCounty govts capacity and
programs have at best no effects --at worst, detrimental effects on
job growth, poverty rates, and household income.
IV. Empirical AnalysisData: Counties from all 46 contiguous
states with functioning county govts, primary data collected from
62% of counties, along with secondary data (N=1514)Modeling
Relationships follows poverty literatureControl for spatial
dependency with a spatial lag model in a GMM approach.Models seek
to minimize endogeneity (e.g. job growth estimated with
instrumental variable, industry mix)
Partridge--LSU19IV. Empirical AnalysisModeling
RelationshipsExamine four indicators of economic well-being: job
growth (2001-2007) individual poverty rate (2007)poverty rate of
children under age 18 (2007) household median income (2007)Base
modelsrobustness checksStandard diagnostics (no high
multicollinearity, etc.)Partridge--LSU20
Partridge--LSU212007 Poverty Rate by CountyIV. Empirical
AnalysisIndependent Variables (base-model controls for county
conditions)1.Local Economic Structure (1990s)Percent
manufacturingAverage job growth of contiguous counties Government
as a local employer (federal, state, and local)2. County
Demographic Characteristics (1990s) Percent of population: < age
18 and age 65 years and > Percent of black and Latino
populations Education: age 25 with a college degree or +%
Female-headed households with children under age 183. Agglomeration
Log of population densityDistance to nearest MSA in 2000Poverty
pocket: percent of 8 nearest counties with 1997 poverty rate >
20%County 2000 population if county is in an MSA or nearest MSA
otherwise.Distance to reach a large MSA (population at least
250,000)
Partridge--LSU22IV. Empirical AnalysisDependent County Govt
Variables: County Capacity Administrative resources: --County
government size (full-time employees)--Economic development
professional on staff --Grant writer on staffFiscal
resources:--General revenue/general expenditures (higher=greater
fiscal capacity) --Revenue per capita Decentralizationcounty
autonomy: --State and federal/own source revenue (lower=greater
fiscal autonomy)--# of governments operating in county (lower=less
fragmentation, stronger county) --Devolutioncounty located in state
that devolves welfare to counties
Partridge--LSU23IV. Empirical AnalysisDependent Variables:
County Government PolicyThree Contrasting Types of Development
Policy that are debated--Traditional business attraction: (7 policy
tools, e. g. tax abatements) --Alternative (new wave) business
economic development: (8 policy tools, e.g. business incubators,
new technologies) --Programs directed to local workforce: (10 item
community support index e.g., workforce development programs for
low-income workers, childcare services)Additional Measures: County
Economic Development Focus --Business incentives (13 incentives,
e.g. local designated enterprise zones, low-cost loans)
--Proportion of economic development budget for small business
development (index)--County-created industrial
parkPartridge--LSU24IV. Empirical Analysis: ResultsJob Growth: is
positively associated with:County government capacity is larger:
Larger governmentsMore centralized county govts (less fragmented by
many local govts) More autonomous (less dependence on state and
federal funds) These findings tend to support the position outlined
about beneficial effects of stronger county govts. Note: higher
revenue extracted per capita, lower job growth (suggests taxation
dampers growth)Policy: Where a larger number of new wave policies
are in use-- providing an indication they may support job growth.
But where the county provides fewer public services for the
workforce Note: no effects of: traditional, business attraction;
industrial incentives; industrial park; or spending focused on
small business development
Partridge--LSU25IV. Empirical Analysis: ResultsGeneral Poverty
Rate are negatively related to:Capacity is greater: Staff: economic
development professional on staff Fiscal capacity: higher per
capita revenue (control for economic structure) Greater autonomy:
less dependence on state/federal funds; welfare devolved to
countyThese findings tend to support the position outlined about
beneficial effects of stronger county govts. Little effect of
policy:Except a somewhat positive effect of new wave policies
(p