Poverty in Maine - 2006
Introduction

Introduction

This report presents a county-by-county picture of poverty in Maine, which updates and adds to information originally published in the report, Poverty in Maine: 2003.¹ In 2002-2003, the Maine Community Action Association contracted with the Margaret Chase Smith Policy Center at the University of Maine to design and conduct a statewide needs assessment as part of the requirements of the Community Services Block Grant. The Poverty in Maine: 2003 report was part of that project. In 2005-2006, the Community Action Association again contracted with the Margaret Chase Smith Policy Center to provide an update on the 2003 report. The intent of both the earlier report and the current one here is to build upon the Maine State Planning Office’s "Annual Report Card on Poverty in Maine," which looks at indicators statewide, in order to provide a more detailed examination of poverty at the county and local levels. This written report represents a subset of the data compiled for this project. We also have produced a database with additional information at the municipal level, which will be provided to the community action agencies for their use in future needs assessment, planning, and evaluation activities.

Methodology and Data Sources

We have prepared this picture of poverty in Maine by using a few sets of data selected for their measurement properties of accuracy, completeness, and longitudinal availability, rather than using a larger variety of less thorough datasets. The datasets selected for analysis are from the food stamp program administered by the Maine Department of Health and Human Services, the Low Income Home Energy Assistance Program (LIHEAP) administered by the Maine State Housing Authority (MSHA) through the community action agencies, unemployment data from the Maine Department of Labor, and relevant information from the U.S. Census; the updated Census information on poverty and income reported here is from the U.S. Census Small Area Income and Poverty Estimates (SAIPE) program.²

In this report, we provide additional information not included in the 2003 study from several other data sources: (1) information from the federal Bureau of Economic Analysis Regional Economic Information System (REIS), which has detailed data on personal income at the county, state,
and national levels, derived from a variety of sources using a complex methodology; (2) data developed by the Maine Center for Economic Policy (MECEP) regarding living costs and livable wages in Maine’s counties and municipalities (Pohlmann and St. John 2005); (3) data on free and reduced lunch eligibility from the Maine Department of Education.

The datasets used in this report are from slightly different time periods. Information about food stamps and LIHEAP is from the most recent federal fiscal year (October 2004-September 2005). Information on the school lunch program is reported as of October 31, 2005, and is for federal fiscal year 2005-06. Unemployment rates are an average for January through December 2005. "Livable wage" information is for 2004. The most recent data available from the federal Bureau of Economic Analysis REIS is for 2004. The most recent yearly county-level data from the U.S. Census SAIPE program is for 2003.³

Information that is gathered in program implementation is rarely perfectly suited for outcome measurement or for needs assessment. As policy researchers, we almost always work with information that was collected for a different purpose than the task at hand. In social service programs, such as LIHEAP and the food stamp and free and reduced school lunch programs, information is usually collected to establish individual eligibility, avert fraud, and count services rendered.

Our intent, as in the 2003 report, is to use information not just to document what has been done, but to help estimate what has not been done. To gauge unmet need, participation rates for various benefit and assistance programs can be measured against each other and against other measures, such as those from the Census and from the Bureau of Economic Analysis REIS.

In this effort, we have worked interactively with the Maine Community Action Association agency directors to select, analyze, and portray those indicators that are most useful and relevant to the mission of the community action programs and the needs of Maine’s population. The indicators used here are a subset of standard ones used in the United States in evaluating the extent of poverty and assessing needs. These include several types of income measures, poverty rate, employment, and measures of services and benefits aimed at low-income populations.

Report Organization

This report is divided into three sections. In this first, introductory section we give a general description of poverty, income, and benefits indicators. In the second section we provide an overall statewide view of a subset of these indicators. These are presented in a series of maps, charts, and tables with accompanying text. These maps, charts, and tables allow for county-by-county comparisons of the selected indicators, as well as comparison of the county-level information with that for the state as a whole. In this section, we also include a discussion of trends in these indicators since the 2003 report.

The third section provides "poverty profiles" of each county. Each profile includes a series of tables of poverty indicators, a brief narrative analysis of highlights of those indicators, a map for one selected indicator (households in each town receiving LIHEAP assistance in 2004-2005), a chart of three key indicators for the county compared with statewide totals (individuals below poverty level, households receiving LIHEAP, households receiving food stamps), a pie chart showing total personal income by source for the county, and several graphs showing indicator trends for poverty and economic distress from the time of the earlier report to the current period.

Defining and Analyzing Poverty, Economic Distress, and Income

Poverty is a complex, multidimensional concept. Poverty can be chronic or temporary. Broad definitions of poverty might include components such as household income/consumption (e.g., poverty "lines" or thresholds); human capabilities (e.g., education, child nutrition, low birth-weight babies); access to public services (e.g., schools, transportation, health services, safe water and sanitation facilities); employment and assets (e.g., employment rates, housing). Using non-income measures of poverty can provide a more complete assessment of poverty, but it complicates analysis since certain groups that would be considered "poor" by some indicators would not be by others (Lok-Dessallien n.d.).

In the United States, the most widely known and commonly used poverty indicator is the federal poverty measure.⁴ This income-based measure was officially established in 1969 by the Office of Management and Budget, based on work done during the 1950s by Mollie Orshansky, an analyst with the Social Security Administration. Gross cash income for the household is compared with the appropriate threshold and adjusted for family size to determine poverty status. There are two slightly different versions of the federal poverty measure: the poverty thresholds and the poverty guidelines. Both of these are updated annually for price changes using the consumer price index for all urban consumers (CPI-U).

• Poverty thresholds: This is the statistical version of the poverty measure, issued by the Census Bureau. It is used in calculating the number of persons and households in poverty in the United States or in states and regions. The Census poverty threshold uses separate figures for aged and non-aged, one-person and two-person households. In this report, when we refer to households or individuals as being below or above poverty, we are normally using the Census poverty thresholds.
  

• Poverty guidelines: This is the administrative version of the poverty measure, and is issued by the U.S. Department of Health and Human Services (HHS). The poverty guidelines are a simplification of the poverty thresholds and are used in determining financial eligibility for many federally funded programs. The poverty guidelines do not make a distinction between elderly and non-elderly households as do the Census poverty thresholds. Some programs use a percentage multiple of the guidelines in determining eligibility, such as 125%, 150%, or 185%. A major reason for having poverty guidelines distinct from thresholds is that thresholds for a given year are not published in final form until late summer of the following calendar year. The poverty guidelines are sometimes loosely referred to as the "federal poverty level."

Some examples of federal programs that use poverty guidelines in determining benefit eligibility are:

• In the Department of Health and Human Services: Community Services Block Grant, Head Start, Low-Income Home Energy Assistance Program (LIHEAP), Children’s Health Insurance Program
  

• In the Department of Agriculture: food stamps, Special Supplemental Nutrition Program for Women, Infants, and Children (WIC), the national school lunch and school breakfast programs
  

• In the Department of Energy: weatherization assistance
  

• In the Department of Labor: Job Corps, Senior Community Service Employment Program, National Farmworker Jobs Program
  

• In the Legal Services Corporation: legal services for the poor

Some relatively recent provisions of Medicaid use the poverty guidelines, but the rest of that program (accounting for about three-quarters of Medicaid eligibility determinations) does not. A number of the major means-tested programs do not use the poverty guidelines in determining eligibility, including Temporary Assistance for Needy Families (TANF) (and its predecessor, Aid to Families with Dependent Children [AFDC]), Supplemental Security Income (SSI), the Earned Income Tax Credit (EITC) program, the Department of Housing and Urban Development’s means-tested housing assistance programs, and the Social Services Block Grant. Some state and local governments use the federal poverty guidelines in some of their own programs and activities, as do some private companies in determining eligibility for their services to low-income people.

Information on poverty in the Census is derived from a sample of the population, with figures projected for the general population. Poverty status
at the household level is determined based on overall household income reported by respondents (from all cash sources including wages,
self-employment, "social welfare" cash benefits, interest and dividends, and pensions), adjusted for household size and age. Poverty on the individual level is defined as any individual living in a household that is below poverty.

The federal poverty measure has come under a good deal of criticism, and there are ongoing efforts to modify the way the measure is calculated to make it more relevant and meaningful. When the measure was originally developed, food costs accounted for about one-third of household budgets, and the poverty level was calculated by using the cost of a minimum food budget, as determined by the U.S. Department of Agriculture, and multiplying that figure by three. The formula has not been changed, even though food constitutes a much smaller proportion of household budgets than it did when the poverty measure was developed.

Changes in federal policy, regional differences, and changing levels or patterns of consumption by American households have not been incorporated into the federal poverty measure. In terms of policy, changes in the tax code (e.g., increased payroll and income taxes) have changed the amount of available income for households; in-kind benefits (for example, food stamps, housing assistance) are not included in calculations of household resources. Regional variations in the cost of living, especially housing costs, are not considered when determining a household’s consumption needs. Costs of child care, medical care, and health insurance also are not included.

Several important studies and reports have suggested ways in which the federal poverty measure can be revised. The Bureau of the Census has issued a series of reports on experimental measures of poverty, so progress toward modifying the federal poverty measure is being made. However, for now, program planning and evaluation and policy studies will continue
to rely on the existing federal poverty thresholds and guidelines.⁵

Income

Three of the most widely used measures of household income are the Census Bureau’s measure of money income, the Internal Revenue Service’s measure of adjusted gross income of individuals, and the Bureau of Economic Analysis’ measure of personal income (BEA, n.d.). Poverty information reported by the Census Bureau is based on self-reporting of money income. Adjusted gross income of individuals as reported to the IRS excludes some categories of income, particularly non-monetary benefits such as food stamps. Economists generally consider the personal income measure to be the most comprehensive of these three income measures.

Personal Income. For both the national and regional levels, personal income includes three broad types of income: net earnings (from wages and self employment); income from investments (dividends, interest and rent); and income from transfer payments. Examining the breakdown
of income from each of these types can tell us a lot about the economic characteristics of an area. For example, having a higher proportion of personal income from transfer payments in a given state, region, or county is generally an indicator of higher poverty levels, presence of an older population, or both combined.

Transfer payments are payments for which no current services are performed. These are payments by federal, state, and local governments and by businesses.⁶ Government payments to individuals include

• retirement and disability insurance benefits (e.g., Social Security [old age, survivors’, and disability benefits], worker’s compensation),

• medical payments (mainly Medicare and Medicaid),

• income maintenance benefits (e.g., TANF, food stamps, SSI),

• unemployment insurance benefits,

• veterans’ benefits, and

• federal grants and loans to students.

Transfer payments may be made directly to individuals (e.g., retirement and disability insurance payments, income maintenance benefits, unemployment insurance benefits), or they may be made on behalf of individuals (e.g., medical payments—Medicare, Medicaid—paid to providers). Because personal income amounts include government medical benefits, per capita personal income figures are higher than per capita income as computed by the Census’ measure of money income or the IRS’ adjusted gross income figures.

Some types of transfer payments are means-tested, that is, they are based on income level formulas. These include income maintenance benefits such as TANF, food stamps, and SSI, as well as medical payments to providers for most Medicaid programs. However, most transfer payments are non-means-tested, and are sometimes colloquially referred to as "entitlements." These include government retirement and disability benefits (e.g., Social Security and military pensions), unemployment compensation, Medicare payments to providers, and some Medicaid payments (e.g., for the disabled).

1. Some of the general background material from the 2003 report is included verbatim in this current report, since it is still relevant and is important to understanding how poverty is measured and why particular indicators are being used.

2. The U.S. Census Bureau, with support from other federal agencies, created the Small Area Income and Poverty Estimates (SAIPE) program to provide more current estimates of selected income and poverty statistics than the most recent decennial census. Estimates are created for states, counties, and school districts. The main objective of this program is to provide updated estimates of income and poverty statistics for the administration of federal programs and the allocation of federal funds to local jurisdictions. These estimates are derived from small samples, not from surveys of the entire population which are only done every 10 years.

3. County-level data files on personal income for 2004 were released by the Bureau of Economic Analysis on April 25, 2006. County-level poverty data from the Census for 2003 were released in November 2005, and data for 2004 is expected to be released in November 2006.

4. Information here on the federal poverty measure and on programs using and not using the federal poverty guidelines is derived from the excellent University of Wisconsin Institute for Research on Poverty Web site, in the "Frequently Asked Questions" section, http://www.irp.wisc.edu/faqs.htm

5. A useful summary of a 1999 conference evaluating the federal poverty measure may be found in a paper by Thomas Corbett, "Poverty: Improving the Measure after Thirty Years: A Conference," which is available on the University of Wisconsin Institute for Research on Poverty Web site at http://www.irp.wisc.edu/publications/focus/pdfs/foc202.pdf#page=51. Links to the Census Bureau’s reports on experimental poverty measures may be found at http://www.census.gov/hhes/poverty/povmeas/reports.html

6. Business payments to individuals consist primarily of liability payments for personal injury, which represent only a very small proportion of total transfer payments.

Previous: Executive Summary