• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Conceptualization and measurement of poverty: A comparative analysis

Dissertation
Author: Moses Kwadzo
Abstract:
Many poverty authors point out that the various ways poverty is conceptualized and measured are very crucial because different poverty measures tend to capture different people as poor. The main focus of this research is to compare and examine how different poverty measures estimate poverty outcomes in the United States. The data for the study are from the 2004 of National Longitudinal Study of Youth, 1979 (US Department of Labor, 2006). Using frequency curves and cross tabulations, the distribution of the sample in poverty was computed. Logit models were used to estimate the likelihood of the effects of demographic characteristics on individuals to fall below the thresholds of three poverty measures that include the monetary, social exclusion, and capability poverty measures. All the three poverty measures have been found to estimate varied levels of poverty outcomes. Both the monetary and social exclusion poverty measures are found to exhibit consistent patterns in the distribution of sample in poverty. On the contrary, poverty incident is found to be highly sensitive to the capability poverty lines. In addition, the research findings also indicate that the poverty measures do overlap to capture a percent of the sample as poor. The result indicates a moderate agreement between the monetary and social exclusion poverty measures. However, the consistency between the capability and monetary or social exclusion poverty shows a weak agreement. The findings that the capability poverty measure exhibits inconsistent patterns of poverty distribution as well as a weak agreement with the other two poverty measures suggests that the capability poverty measure is not a good poverty measure. The logit analyses have also shown that gender (being female), race (other than White), place of residence (rural or inner city dwellers), and marital status (never married, separated, divorced, widowed) had statistically significant positive effects on the likelihood of individuals falling into poverty. Of these, marital status is the strongest predictor in determining the likelihood of persons falling into poverty.

TABLE OF CONTENTS

CHAPTER Page

ONE INTRODUCTION……………………………………………… ……….1

United States’ Welfare Policies and Poverty Reduction ………………...1 History of Poverty Thresholds……………………………………………8 Statement of the Problem……………………………………………......12 Significant of the Study………………………………………………….14 Definition of Terms……………………………………………………...15 Monetary Poverty……………………………………………………16 Capability Poverty…………………………………………………...16 Social Exclusion Poverty…………………………………………….17 Participatory Poverty………………………………………………...18 Purpose of the Study……………………………………………………..18 Research Questions………………………………………………………18 Organization of the Study………………………………………………..19

TWO LITERATURE REVIEW………………………………………………..20

Theories of Poverty……………………………………………………...20 Individual Factors……………………………………………………20 Cultural and Neighborhood Factors ……………………………........23 Structure Factors.………………………..…………………………...25 Defining and Measuring Poverty………………………………………...31 Monetary Poverty ……………………………………………………32 Determinants of Monetary Poverty…………………………………..33 Capability Poverty …………………………………………………..35 Determinants of Capability Poverty………………………………….37 Social Exclusion Poverty…………………………………………….40 Determinants of Social Exclusion Poverty…….…………………….43 Conclusion…………………………………………………………...46 Description of the Conceptual Framework….. ………………………….47 Hypotheses: Predicting the Three Poverty Types………………………..52

THREE METHODOLOGY………………………………………………………54

Research Paradigm and Design…………….……………………………54 Data Source.………………………………………………………...........56 Sampling Procedure……………………………………………………...57 Missing Data……………………………………………………………..59 Measurement of Dependent Variables……………………………...........62 Gender and Race……………………………………………………..62 Age……………………………………………………………….......64 Urban/Rural Place of Residence …………………………….............64

viii

Family Size ………………………………………………………….65 Marital Status………………………………………………………...66 Operationalization and Measurement of Criterion Variables……………67 Monetary Poverty (Federal Poverty Thresholds)…………………….67 Capability Poverty (Educational Achievement)……………………..68 Social Exclusion Poverty (Median Income)…………………………72

FOUR RESULTS……………………………………………………………….74

Distribution of Sample Across the Three Poverty Thresholds………….74 Percent in Poverty…………………………………………………...74 Rate of Poverty Distribution………………………………………...76 Economic Participation and Percent in Poverty…………………………79 Cross Tabulation Showing Overlaps Between Poverty Measures………80 Multivariate Analysis……………………………………………………84 Regression Diagnostics ……………………………………………...86 Logistic Coefficients and Odds Ratios ……………………………...90 Relationship between the Monetary Poverty Thresholds and Predictor Variables……………………………………………………….91 Relationship between the Social Exclusion Poverty Thresholds and Predictor Variables……………………………………………………....98 Relationship between the Capability Poverty Thresholds and Predictor Variables……………………………………………………...104 Comparing Significant Predictors of the Different Poverty Measures…110 Poverty Measure and Regional Variations……………………………..113 Summary and Conclusion……………………………………………....117

FIVE DISCUSSION OF RESULTS....…………………….…………….........118

Distribution of Individuals under the Monetary, Social Exclusion and Capability Poverty Measures….…..……………………………………119 Similarities and Differences between the Predictor Variables Associated with the Monetary, Social Exclusion and Capability Poverty Measures…………………………………………...123 Age…………………………………………………….……………124 Level of Family Size………………………………………………..124 Gender………………………………………………………………125 Race…………………………………………………………………126 Marital Status……………………………………………………….127 Urban and Rural Place of Residence……………………………….128 Comparing Odds Ratios for the Predictor Variables……………….129 Policy Implications and Applications…………………………………..131 Limitations of the Study and Suggestions for Further Studies…………133 Conclusion……………………………………………………………...135

ix

REFERENCES…………………………………………………………138

APPENDICES………………………………………………………….151

x

LIST OF TABLES Table

1. Basic Needs and Their Indicators ……………………………………….37

2. Median Earnings (12 months) of Full-time Occupational

Category and Sex………………………………………………………...45

3. Median Earnings (12 months) of Full-time Wage and Salary Workers by Race and Sex……………………………………………….45 4. Characteristics of United States National Population and

Survey Sample…………………………………………………………...61

5. Respondents Race/Ethnicity……………………………………………..63

6. Descriptive Statistics on Rural and Urban Residence……………………65

7. Descriptive Statistics on Marital Status………………………………….66

8. Percent Distribution of Sample in Poverty………………………………75

9. Percent Distribution of Working Sample in Poverty…………………….79

10. Cross Tabulation Showing Overlap Between Poverty Measures………..81

11. Cross Tabulation Showing Overlap Between Poverty Measures………..82

12. Cross Tabulation Showing Overlap Between Poverty Measures………..83

13. Tolerance and VIF among variables (Federal poverty level)……………87

14. Tolerance and VIF among variables (Highest Grade Completed, years)..88

15. Tolerance and VIF among variables (Total Household Income)………...89

16. Logistic Regression of Predictor Variables on Federal Poverty

Threshold (FPT) 1 ……………………………………………………...…94

17. Logistic Regression of Predictor Variables on 1.5 Federal

Poverty Measure 2 ………………………………………………………...95

xi

18. The Observed and Predicted Frequencies for Monetary Poverty (1.5 Federal Poverty Threshold) Logistic Regression………………………96 19. The Observed and Predicted Frequencies for Monetary Poverty (Federal Poverty Threshold) Logistic Regression……………………...97 20. Logistic Regression of Predictor Variables on Median Income(50%) 3 ..100

21. Logistic Regression of Predictor Variables on Median Income(60%) 4 ..101

22. The Observed and Predicted Frequencies for Social Exclusion Poverty (Median Income (50%)) Logistic Regression………………………….102 23. The Observed and Predicted Frequencies for Social Exclusion Poverty (Median Income (60%)) Logistic Regression…………………103 24. Logistic Regression of Predictor Variables on Highest Grade

Completed (Grade 11 or Less) 5 ……………..…………………….…...105

25. Logistic Regression of Predictor Variables on Highest Grade

Completed (Grade 12or Less) 6 ……………………………..…………106

26. The Observed and Predicted Frequencies for Capability Poverty (Equal to or Less than Grade 11)) Logistic Regression………………107 27. The Observed and Predicted Frequencies for Capability Poverty (Equal to or Less than Grade 12)) Logistic Regression………108 28. Regression Models…………………………………………………….109

29. Comparing Significant Predictors of the Different Poverty Measures...112

30. Percent Distribution of Sample in Poverty by Regions………………..114

31. Percent Distribution of Working Sample in Poverty by Regions……...114

32. Comparing Significant Predictors of the Monetary Poverty

xii

for the Four Regions……………………………………………………116

xiii

LIST OF FIGURES Figure

1. Conceptual framework of monetary, capability, and social exclusion poverty…………………………………………………51 2. Distribution of people in monetary poverty……………………………...77 3. Distribution of people in social exclusion poverty ……………………...78 4. Distribution of people in capability poverty…………………………….78

1

CHAPTER ONE INTRODUCTION According to the World Development Report (2001) and World Bank (2002), poverty policies have utilized a broad conceptualization of poverty associated with different dimensions of poverty. Schiller (2008), Laderchi, Saith and Stewart (2003), and Jordan (1996) pointed out that the way we conceptualize and measure poverty influences the fundamentals of poverty policies and programs. While different poverty measures have been utilized, little attention has been paid to their comparative outcomes and implications (Bell, 1995; Blank, 2007; Schiller, 2008). United States’ Welfare Policies and Poverty Reduction Poverty is one of the major social problems in the United States (Hurst, 2004; Rank, 2004). As of 2008, there are approximately 39.1 million, or 13.2%, of Americans in poverty (US Bureau of Census, 2008). Since the advent of industrialization, economic changes led to low earnings, increased social inequality, and increased inability of breadwinners to support their families in many parts of the world, including the USA (Abramowitz, 1996; Trattner, 1994). In order to reduce poverty social institutions and policies are instituted to address the needs of the disadvantaged. Due to differential perspectives regarding causes of poverty, the Federal government’s social policy expanded and contracted government welfare responsibilities toward poverty since 1935 to the early 1980s (Abramowitz, 1996; Trattner, 1994; Woodside & McClaim, 2006). The great historians, Schlesinger (1986) and Schlesinger (1949) postulated that since the civil war the United States experienced spiral swing in social policy between liberalism and conservatism. According to the Schlesingers and Stone (2002), the

2

conservative perspective supports capitalism, private properties, free market and individual initiatives. The liberal perspective is more concerned about collective national effort towards equality, social responsibility and general welfare to ensure decent quality of life. For instance, the periods of Conservative Rule (1869-1901), Republican Restoration (1919-1931), The Eisenhower Era (1952-1960) and Reagan Era (1980-1992) were more absorbed in conservatism (Mandell & Schram, 2003; Schlesinger, 1986; Schlesinger, 1949). These eras concentrated on private property, profit, free market and a reduction in welfare budgets. The Progressive Era (1901-1919), the New Deal (1931- 1947), the Fair Deal (1947-1952), the Great Society (1960), and the Culture of Narcissism (1968-1980) were characterized by liberalism (Mandell & Schram, 2003; Schlesinger, 1986; Schlesinger, 1949). Since the mid-1960s, anti-poverty policies and programs have been implemented in the United States. The Great Depression in the 1930s affected a large number of Americans. Many people were unemployed, a situation that created a difficult time for millions of citizens (Abramowitz, 1996; Mandell Schram, 2003; Trattner, 1994). At the time of the Great Depression the Roosevelt administration implemented several emergency relief programs to address the problems of poverty. The Social Security Act was created in 1935 to meet the general welfare needs of United States citizens. The Social Security Act placed responsibility on the Federal government to take care of the aged, the unemployed, and the disabled. Early Social Security Act programs included Old Age Assistance, Old Age Insurance, Unemployment Compensation, Unemployed Insurance, Public Assistance, Aids to the Blind, and Aid to Dependent Children (Abramowitz, 1996; Trattner, 1994; Woodside & McClaim, 2006). In the mid-20 th

3

century President Kennedy and Johnson’s administrations were committed to the eradication of poverty in the United States. Orshansky developed the official United States poverty measure in 1965. This measure has been used to measure poverty trends and the economic needs of the disadvantaged over the years. The 1960s saw a rapid decline in the poverty rate, while subsequent years indicated fluctuations in poverty rates between 11%-15% with changes in macroeconomic conditions. In 1959, the United States poverty rate was 22.4%. In the 1960s the poverty rate fell steadily and reached as low as 11% in 1973. Since 1973 the United States poverty rate fluctuated between 11%-15% (US Bureau of Census, 2000; Rank, 2001). Piven and Cloward’s (1993) analysis of the poverty in the United States revealed that poverty often decreased during economic expansion and increased during recession. In the early 1980s and 1990s the poverty rate increased as the country experienced economic recession. In the mid-1980s and 1990s, which were periods of economic expansion, the poverty rate decreased (US Bureau of Census, 2000; Rank, 2001). Poverty rate trends in the United States over the past 35 years suggest lack of progress against poverty. Haskins (2006), Hays (2005), Piven and Cloward (1993) and Murray (1994) argued that previous government welfare programs did little or failed to reduce poverty. Murray (1994) and some conservatives noted that welfare encourages poverty and that welfare should be discontinued (Murray, 1994). Charles Murray (1994) presented a historical account and policy analysis of government social programs during the period from 1950 to 1980 and argued that American social policy seemed to advance social problems. He advocated for the dismantling of social programs. Murray believed that

4

government orientation toward capitalism has the tendency to reduce social problems, including poverty. According to Murray, the United States economy that embraced the free market system with minimal government interference from 1950 to 1965 was marked with a high level of poverty reduction from one-third of the population to just over one-tenth. From 1965 to 1980 a number of welfare programs including Aids to families with Dependent Children (AFDC), food stamps, subsidized housing, job programs, and unemployment insurance were instituted to reduce poverty. In his analysis of welfare programs, Murray found positive association between welfare and social problems. He concluded that an increase in welfare programs can result to increased rates of social problems including poverty, marriage breakdown, child illegitimacy and crime. For example, Murray believed the AFDC paid more money to mothers to have illegitimate children and to remain unwed. Proponents of welfare such as Haskins (2006), Hays (2005), Piven and Cloward (1993) point out that those social programs are necessary to reduce social problems such as poverty, yet welfare policy is not effectively crafted. Downs (1998) and Stone (2002) also argued that social policies are sometimes ill-prepared to address real social problems. Anthony Downs (1998) suggested that major social problems go through the issue attention cycle in the policy arena if they possess three specific characteristics. The first characteristic is an association with a minority. Here, the term minority refers to a relatively small number of a population that experiences a given social problem. Although the proportion of population suffering from a social problem is relatively small (e.g.15% of the population), the number of people that suffer from the social problem is large. Secondly, According to Downs, a social problem that impacts minorities is given

5

little policy consideration because the majority of the population does not suffer from that particular social problem. Furthermore, a social problem is often caused by a large social structure or arrangement that benefits a large proportion or a powerful minority of the population. Finally, such a social problem lacks an intrinsically exciting quality which makes it difficult to compete with other issues on public agenda. According to Downs (1998), American public attention or policy does not focus on crucial social problems when they possess the above three features. Downs noted these social problems receive strong public concern in their early stage for a short period and then gradually lose their vitality and finally fade away because such problems are unable to assume enough political momentum in the policy arena. The five stages of Downs’ issue attention cycle are: pre- problem stage, alarm discovery and euphoric enthusiasm stage, cost realization state, gradual decline of public interest, and post problem stage. The pre-problem stage is when some interest groups or experts become alarmed about a highly undesirable social problem. At this point the general public may not be aware of the issue or the problem. The alarm discovery stage may be associated with a series of dramatic events such as riot and social movements that trigger sudden public awareness. According to Downs (1998), the alarm discovery stage tends to move the problem to the second phase, euphoric enthusiasm. At this phase policy makers and society think the problem can be addressed in a relatively short time without giving thought to restructuring the social structure that caused the problem. The third stage of Downs’ issue attention cycle is the cost realization stage. At this stage policy makers, as well as the public, attempt to solve the social problem and gradually come to the

6

realization that the cost of solving the problem is very high. The cost often requires large financial inputs, as well as sacrifice by the large population. Social and structure rearrangement is also needed to solve the problem. Due to the fact that politicians and larger beneficial groups do not want to lose the benefits they enjoy as a result of the existing social structure or arrangement, their interest in solving the problem tends to decline (Downs, 1998). The fourth stage of Downs’ cycle is the gradual decline of intense public interest. At this stage policy makers or other people get discouraged about the cost of solving the social problem following the cost realization. The last stage of Downs’ cycle, the post problem stage, is when the social problem fades away and the problem is replaced by a fresh public issue that is about to move to its crisis or policy agenda stage. The social problem may experience a prolonged limbo of less public attention or recurrence of interest. In order to shift public or policy attention from the social problem, policy makers tend to institute policy programs that do not address the root cause of the social problem (Downs, 1998). Stone (2002) argued that rationality has characterized much of public policy formulation and analysis. As noted by Stone, Allison (1999) and Kingdom (1995), the most commonly used models for policy creation and analysis are the rational actor model and the incremental (instrumentalism) model. The incremental model involves a continuation of an existing government policy program with a minimal changes made to the existing policy. In this case, a systematic periodic review of existing policy based on feedback from the current policy program is often applied. According to Stone, these policy analytical models are based on the assumption of rational choice theory. The

7

concept of rationality emphasizes that an individual tends to be logical, and dispassionate to maximize benefit and to reduce cost (Kahler, 2001). In general, rational decision making occurs in stages that include identification and prioritization of goals, searching for alternatives, predicting the consequences of each alternative, and evaluating each alternative in terms of the set goal and selecting the best alternative with maximum benefit (Allison, 1999; Kahler, 2001; Kingdom; Stone, 2002). The basic assumption of rational decision making is that policy makers have access to all available information on the problem, including alternatives and the consequences of any given policy. Both the rational actor and incremental models rely heavily on the rational decision making process. Policy analysis often rests on rational analysis by which a single criterion is used to measure and judge the policy existence and effectiveness. Stone (2002) pointed out that models that rely on rational decisional making process are unproductive and irresponsive to social problems. Stone argued that policy analysis is an art, given that policy formulation and implementation is a complex task, and that policy makers and analysts do not always have full knowledge of the social problems and their alternative solutions. Moreover, all the actors are not involved in the policy making process. Stone (2002) also argued that the rational actor model fails to capture vital subjective elements that are important in policy analysis and evaluation. Stone underscored those clusters of subjective characteristics that include the conceptualization of social problems, use of metaphors and symbols, the use of language and information distortion, the issue of social cost versus individual cost to resolve social problems, the use of numbers to define and evaluate social problems, and the role of values in policy

8

formulation and analysis. The need to give great consideration to these clusters of subjective features during policy formulation, analysis and evaluation was emphasized. The use of numbers to conceptualize complex social problems can be problematic. For instance, the United States uses headcount measure of poverty to determine the number of people that fall below the poverty line (Bell, 1995; US Department of Health and Human Services, 2007). In this introductory chapter, I first discuss the policy analytical frameworks and historical account of poverty thresholds. I then present the statement of the problem, purpose of the study, and research questions. In the last section of the chapter, the significance of the study, definition of terms, and the organization of the next chapters are presented. History of Poverty Thresholds The United States poverty measure utilizes an income-based approach that defines poverty as a minimum amount of household income below the poverty line. There are two slightly different versions of the United States federal poverty measure, the Poverty Thresholds and Poverty Guidelines (US Department of Health and Human Services, 2009 & 2007). The poverty threshold is a set of money income thresholds that vary by family size and composition (based on ages and number of family members). Any family or individual whose total income falls short of the poverty threshold is regarded poor. The Census Bureau uses the same poverty threshold throughout the USA to determine poverty status. (US Department of Health and Human Services, 2009 & 2007). The poverty guidelines are the second version of the federal poverty measure, which are released each year by the Department of Health and Human Services (HHS).

9

They are a simplified version (percentage multiples such as 150% or 1.5 ) of poverty thresholds that is used for administrative purposes by many government aid programs to determine the individuals’ eligibility for certain federal programs (US Department of Health and Human Services, 2009 & 2007). The poverty guidelines do not vary geographically and they tend to be the same nationwide. Government aid programs do not usually use the poverty thresholds to determine their client’s eligibility for public assistance. While the poverty thresholds are used to calculate the number of people in poverty, the poverty guidelines are used to determine people’s eligibility for certain programs. Different percentage multiples of the poverty guidelines may be used to determine individuals’ eligibility for programs. Some of the government aid programs use the simplified version of poverty thresholds, poverty guidelines, to determine individuals’ eligibility for certain Federal programs including Head Start, the Food Stamp Program, the National School Lunch Program, the Low-Income Home Energy Assistance Program and Children’s Health Insurance Program (US Department of Health and Human Services, 2007). On the other hand, the poverty guideline is adjusted to reflect the price of the calendar year (US Department of Health and Human Services, 2009 & 2007). The history of this poverty measure can be traced to the work of Orshansky in the early 1960s (Bell, 1995; Fisher, 1992; Schiller, 2008; US Department of Health and Human Services, 2007; Weinberg, 1994). Mollie Orshansky first formulated the United States’ poverty thresholds. Before Orshansky developed the poverty measure, the United States did not have any single officially acceptable definition and/or measurement of poverty. From 1899 up to 1946, the United States’ measurement was based on "standard budgets”, goods and services; a family of a certain size would need to live at a certain

10

level of well-being. In the early 1900s, the poverty line was below $460 for a five-person family in the North and below $300 for the same size family in the South. In 1949, a study commission by the Congressional Joint Committee on the Economic Report recommended the poverty line be $1,000 for farm families and $2,000 for nonfarm families. In the later years, the official poverty line was estimated and published from the 1960’s US Bureau of Census and Current Population Survey based on the assumption that both small and large families had the same income level (Bell. 1995; Fisher, 1992; Weinberg, 1994). Orshansky developed the United State’s official poverty threshold at the time President Johnson declared war on poverty in the country. Orshansky, an economist who worked for the Social Security Administration (SSA), was initially interested in a project that became “Children of the Poor.” Orshansky wanted to assess the relative risks of low income families with children (Bell, 1995; 1994; Fisher, 1992; Weinberg, 1994). Orshansky used data from a 1955 Household Food Consumption Survey. She initially developed two sets of poverty thresholds. One threshold was developed based on the Agriculture Department’s economic food plan, and the second poverty threshold was developed using the Agriculture Department’s low-cost food plan. The low-cost food plan related to food patterns of families in the lowest third of the income range while the economy food plan cost about 75%-80% as much as the low-cost food plan. By computing food-cost to total expenditure, Orshansky estimated that food expenditure represented about one-third of total family income. Orshansky’s poverty threshold is a combination of both absolute and relative definitions of poverty. She therefore described poverty as a “relative absolute” measure of poverty (Bell, 1995; Fisher, 1992; Weinberg,

11

1994). The absolute poverty thresholds are fixed at a point in time and are updated solely for price changes. On the other hand the relative poverty thresholds are developed by reference to the actual expenditures (or income) of the population. Orshansky defined poverty as more than one-third of total family income spent on food (Bell, 1995; Fisher, 1992; Weinberg, 1994). The Office of Economic Planning adopted Orshansky’s poverty thresholds in 1965 as a working definition of poverty for statistical, planning and budget purposes. The official poverty thresholds vary by family size and composition, but it does not vary geographically (Bell, 1995; Fisher, 1992; Weinberg, 1994). It is used to estimate the number of Americans in poverty annually (US Census Bureau, n.d.; US Department of Health and Human Services, 2007). Thus, the poverty threshold is a set of income indicators that varies by family composition (based on ages and number of family members), and any family or individual whose total income falls short of the poverty threshold is regarded poor (US Census Bureau, n.d.; US Department of Health and Human Services, 2007). For example, the 2006 poverty threshold indicated the following poverty thresholds: for one person under 65 years the poverty line was $10,488, for a two-person family (person under 65 years with a child under 18) poverty line was $13,500 and for a three-person family (two persons under 65 with a child under 18) poverty line was $16,227 (US Department of Health and Human Services, 2007).

12

Statement of the Problem Poverty and social policy research utilize a diversity of poverty definitions and measurements. A general examination of poverty policy and program evaluation studies reveals different conceptualizations and measurements of poverty and well-being. The use of different poverty measurements has been associated with different and/or contradictory outcomes (Bell, 1995; Laderchi et al., 2003). However, the choice of a particular definition and measurement of poverty has important consequences for the poor. According to Laderchi et al.(2003), the various conceptualizations and measurements of poverty may not point to same people as being poor, and this could lead to a particular policy consideration for poverty reduction. Laderchi et al (2003). and Hagenaars and Vos (1988) emphasized that the choice of a specific definition and measurement of poverty may result in different estimates of the determinants of poverty and evaluation outcomes for poverty programs. However, researchers and policymakers often prefer to adopt a particular definition of poverty based on the availability of data, political interest or historical justification (Behn, 1995, 2003). While the choice of a specific poverty indicator may have major consequences for poverty reduction, some indicators may be a better measure for a specific poverty situation (Hagenaars & Vos, 1988; Laderchi et al, 2003). In particular, United States poverty policy and program evaluation outcomes are associated with different estimates of the poor and/or contradictory outcomes. This could be attributed to differences in the conceptualization and measurement of poverty and well-being. While some evaluation studies mainly focus on indicators of monetary poverty, other studies utilize either capability or social exclusion indicators to support

13

their findings (Hagenaars & Vos 1988; Laderchi et al., 2003). The rate of poverty is not only an important social indicator of well-being of the poor, it provides a useful tool that can shape social policy and target programs that benefit the poor. While the American dream emphasizes a wide range of economic outcomes including educational achievement, good health, proper housing and material possessions, the United States official poverty measure focuses on the use of an absolute poverty line (Blank, 2007; Merton, 1968, 1938). Although Orshansky’s 1965 poverty definition is fundamental to identifying poverty in the United States, it has been criticized as being far from the reality of the phenomenon and experience of poverty in contemporary society. Many socioeconomic changes have occurred over the years and the United States official poverty thresholds fail to capture those factors that affect families’ poverty and well-being (Fisher, 1992; U.S. Bureau of Census, 2004). The official poverty measure is noted to have both methodological and resource definition flaws (Blank, 2007; Dalaker, 2005). Methodologically, Orshansky estimated that food expenditure represented about one-third of total family income. Thus, the official poverty measure was calculated as three times subsistence food budget. In today’s economy, less than one-third of total income is spent on food. A poverty threshold based on a simple commodity is inappropriate because it makes the threshold numbers more sensitive to the price of that food than the price of any other expenditure for low-income families. Blank argued that an increase in income or change in behavioral expectation will require changes in those things needed for an individual or families to fully participate in society. Blank also noted that the United States official poverty measure fails to fully capture low-income families’ material deprivation.

Full document contains 172 pages
Abstract: Many poverty authors point out that the various ways poverty is conceptualized and measured are very crucial because different poverty measures tend to capture different people as poor. The main focus of this research is to compare and examine how different poverty measures estimate poverty outcomes in the United States. The data for the study are from the 2004 of National Longitudinal Study of Youth, 1979 (US Department of Labor, 2006). Using frequency curves and cross tabulations, the distribution of the sample in poverty was computed. Logit models were used to estimate the likelihood of the effects of demographic characteristics on individuals to fall below the thresholds of three poverty measures that include the monetary, social exclusion, and capability poverty measures. All the three poverty measures have been found to estimate varied levels of poverty outcomes. Both the monetary and social exclusion poverty measures are found to exhibit consistent patterns in the distribution of sample in poverty. On the contrary, poverty incident is found to be highly sensitive to the capability poverty lines. In addition, the research findings also indicate that the poverty measures do overlap to capture a percent of the sample as poor. The result indicates a moderate agreement between the monetary and social exclusion poverty measures. However, the consistency between the capability and monetary or social exclusion poverty shows a weak agreement. The findings that the capability poverty measure exhibits inconsistent patterns of poverty distribution as well as a weak agreement with the other two poverty measures suggests that the capability poverty measure is not a good poverty measure. The logit analyses have also shown that gender (being female), race (other than White), place of residence (rural or inner city dwellers), and marital status (never married, separated, divorced, widowed) had statistically significant positive effects on the likelihood of individuals falling into poverty. Of these, marital status is the strongest predictor in determining the likelihood of persons falling into poverty.