Public Affairs

Permanent URI for this collectionhttps://digital.lib.washington.edu/handle/1773/19501

Browse

Now showing 1 - 20 of 41

Investing in Wellbeing: Three Studies on Public Benefits, Family Processes, and Mental Health in the US
(2025-10-02) Lindman, Tom; Hill, Heather D
My dissertation includes three studies that examine whether public benefits affect the mental health of parents and kids. The first two studies evaluate whether the Community Eligibility Provision—which expands access to free school breakfast and lunch—affects family processes, child behavioral outcomes, parent mental distress, and child mental health diagnoses. The third study evaluates whether Washington's Paid Family and Medical Leave Program affects depression and anxiety among mothers following the birth of a child. All three studies shed light on whether U.S. social safety net programs affect mental health outcomes. In Chapter I, I investigate whether expansions in access to free school meals affect measures of child and parent wellbeing. The Community Eligibility Provision (CEP) is a federal policy passed in 2010 allowing US schools and districts serving a high proportion of low-income families to provide free breakfast and lunch to all students. Eligible schools that opt into the CEP must provide free meals to all students at the school and stop tracking student-level free meal eligibility. Beyond the direct effects of the CEP on child food security, the policy may act as an in-kind transfer to families, and, in doing so, affect both parent and child wellbeing. Using the Early Childhood Longitudinal Study, Kindergarten Class of 2010–11 (N=4,400) and a difference-in-differences design, this study estimated the effect of CEP exposure on measures of parent wellbeing, investments in children, and child behavioral outcomes among a national cohort of families whose children attended CEP-eligible elementary schools during school years 2010-11 through 2015-16. Overall, I found null effects of CEP exposure on these outcomes, although CEP exposure was associated with a small increase in parent mental distress for higher-income families in the sample. Coupled with the existing CEP literature, results highlight the potential for universal free school meal programs to occupy a unique place in the social safety net, complementing other policies by supporting child nutrition at school with modest impacts on household budgets. In Chapter II, I investigate whether expansions in free school meals via the CEP affect child mental health diagnoses. The diagnostic prevalence of ADHD, anxiety, and depression diagnoses among K-12 students in the U.S. has increased over recent decades, exposing mental health challenges faced by children and youth. As a result, supporting child mental wellbeing is increasingly a priority for public health agencies, schools, and policy makers. Despite evidence that income instability and low socioeconomic status negatively influence child mental health, little research assesses whether U.S. food assistance programs—which provide meals and food subsidies to low-income families—affect the symptoms or diagnosis of child mental health conditions. I evaluate whether implementing universal free school meals impacts the likelihood of child ADHD, anxiety, and depression diagnoses among a population of children visiting Federally Qualified Health Centers. My identification strategy leverages the rollout of the Community Eligibility Provision using a difference-in-differences design which allows for staggered policy adoption. Results indicate that CEP exposure is associated with a 3.3% (0.34 percentage point) decrease in the likelihood of ADHD diagnosis, not associated with anxiety diagnosis, and associated with a 9.0% (0.27 percentage point) decrease in the likelihood of depression diagnosis for male students. In Chapter III, I investigate whether paid family leave affects the likelihood of new mothers being treated for depression or anxiety following the birth of a child. Research indicates that when workers have paid time off to care for themselves and others, they have more stable economic circumstances and improved physical health. However, less is known about the effect of access to paid leave on mental health, particularly among parents using leave around a birth. Building on the existing paid leave literature, I evaluate the impact of paid family leave on the likelihood of parents receiving care for depression and anxiety in the quarter following childbirth. My study leverages a quasi-random feature of Washington's Paid Family and Medical Leave eligibility rules—the work-hour eligibility threshold— using regression discontinuity designs to estimate the policy's impact on the likelihood of parents submitting depression and anxiety-related insurance claims. My dataset links paid leave administrative data from Washington with medical insurance claims from the WA All Payer Claims Database. Results suggest that access to Washington's Paid Family and Medical Leave program reduces the likelihood of mothers submitting medical claims related to depression by approximately 26.5%, with larger impacts among leave users. I find no discernable impact on insurance claims related to anxiety. This study is the first in the U.S. to evaluate paid leave's impact on measures of mental health care or diagnosis and speaks directly to the efficacy of state paid leave programs at supporting the wellbeing of working parents. All three studies speak to social policy's effect on parent and child mental health. My research questions address the efficacy of social safety net programs at supporting the health and wellbeing of working families, while adding to our understanding of how investments in families might produce economic benefits over the medium or long term. To motivate my research questions and analyses, I draw on concepts from fields including child development, social epidemiology, demography, and labor economics. I use novel data sources including administrative data, insurance claims, and electronic medical records coupled with causal inference methods. I apply an interdisciplinary perspective to this work in order to motivate causally focused research on how public benefits affect parents, children, and interaction within families, with the goal of producing rigorous social science that speaks to current policy debates.
The Cost of Being Trans: Administrative Burden, Citizen-State Interaction, and Transgender People in the US
(2025-05-12) Sederbaum, Isaac; Martin, Karin D
This dissertation explores how transgender people in the US experience administrative burdens. Across three chapters, I explore the learning, compliance, and psychological costs that trans and gender-expansive people face when applying for safety net programs, interrogating how policymakers often use norms to create burdens, which in turn create and reify norms further. I also look at how trans and gender-expansive people cope interacting with government workers, asking, when do burdens become sources of violence? Employing a mixed-method approach, I first survey 465 transgender adults residing in the US who considered applying or applied to SNAP, Medicaid, and/or Unemployment Insurance. I then interview a subset of 43 survey participants to better understand their experiences of administrative burdens during citizen-state interactions. Taken together, this mixed-method dissertation offers one of the first in-depth studies of transgender Americans' experiences of administrative burden. The first goal of this dissertation is to build a foundation for the field of public administration to better understand what gender is and how it functions. Ultimately, my aim for this dissertation is that it inspires future administrative burden and public administration scholars to understand that gender is not only an independent variable, but also an ever-changing label that can determine who is counted, who is deserving, and who is pushed to the margins.
Understanding and Addressing the Persistence of Poverty in Later Life
(2024-09-09) Freitag, Callie; Hill, Heather D.
Poverty among older adults in the United States is persistent and puzzling. Older adults in the United States are often referred to as the “deserving poor” because they are afforded a near-universal and more generous safety net than younger adults. Yet, the safety net falls short of eliminating poverty in later life. In this dissertation, I explore three mechanisms that may contribute to the persistence of poverty in later life despite the relatively well-developed safety net for older adults: exiting the labor force, age-based disability determination rules for Supplemental Security Income (SSI), and the overall design of the SSI program. I take a mixed-methods approach across three papers, answering my questions with data from surveys, state agencies, and qualitative interviews. Across all studies, I define “later life” broadly, usually beginning at age 50, to capture differences in the safety net afforded to adults who reach retirement age. In Chapter 1, I construct longitudinal one-year panels of older workers from Current Population Survey data to assess how the relationship between exiting the labor force and entering poverty changed after the onset of the COVID-19 pandemic. I find that the probability of entering poverty is much higher for older workers who leave the labor force than those who remain. I also find that the likelihood of entering poverty among labor force leavers significantly decreased by 3 percentage points in the first year of the pandemic, but this decrease was consistent with pre-pandemic trends. In Chapter 2, I use administrative microdata from multiple agencies in Washington state to explore pathways to SSI take-up by age and how the Social Security Administration’s age-based disability regulations affect SSI take-up rates. I find significant and discontinuous increases in SSI take-up at age 55 and age 65, which suggests that the age threshold may be arbitrarily delaying otherwise-eligible people from receiving SSI benefits. Additionally, I find that more SSI recipients had experienced homelessness (40 percent) in the five years prior to SSI take-up than had been employed (34 percent). The findings from this research raise policy questions about the timeliness of SSI and the definitions of disability used to determine SSI eligibility. In Chapter 3, I use novel, nationally representative qualitative interview data from the American Voices Project to ask questions about how older SSI recipients describe the pathways that led them to receive SSI, the strategies and resources they use to get by financially given program restrictions, and how they make meaning of the role SSI plays in their financial lives. I find most older SSI recipients have a meaningful history of work, and that SSI’s low benefit levels and program rules keep recipients struggling financially.
Primary Care Physicians as Street-Level Bureaucrats
(2024-09-09) Gellner, Brenda; Hill, Heather D
This dissertation is a theoretical and empirical investigation of primary care physicians as street-level bureaucrats. In chapter one, I provide a brief overview of the dissertation and its contributions to the relevant literature. In chapter two, I establish the theoretical framework used in this dissertation by linking literature from public administration and health services, discussing policy implementation from the top-down and bottom-up perspectives, and the role of administrative discretion in these processes. Furthermore, I review the history of decision-making models in the US health care system, from paternalism to shared decision-making, and then discuss common barriers to implementing shared decision-making and the relevance of discretion in this space. In chapter three, I present findings from in-depth interviews with thirty primary care physicians that focus on perceived discretion. While participants initially perceive themselves to possess high levels of discretion, I identify five common threats to their discretion: (1) the adversarial health care “machine”, (2) difficult and time-consuming insurance interactions, (3) unattainable patient expectations, (4) high emotional labor, and (5) the undervaluation of primary care. Participants use multiple coping mechanisms to mitigate these threats, with varying success. This chapter contributes to a richer understanding of perceived discretion and its relationship with patient-centered care and physician burnout. In chapter four, I present findings from a vignette-based experiment with two-hundred primary care physicians that focuses on how they mentally sort patients based on perceived engagement and urgency, and whether these perceptions influence their promotion of shared decision-making with these patients. I find that respondents devoted more time in a fifteen-minute appointment to shared decision-making tasks that precede a decision with a fictional high (versus low) A1C patient. Yet, they were less likely to promote a shared decision with that patient or perceive them to be a good candidate for shared decision-making. Perceived patient engagement and urgency played a mediating role in this decision-making process. This chapter provides an additional explanation for why shared decision-making is not occurring as intended. In chapter five, I provide a summary of the dissertation. In summary, this dissertation emphasizes the importance of viewing primary care physicians as street-level bureaucrats. This perspective contributes to our understanding of policy implementation by focusing on primary care physician discretion, coping, and promotion of shared decision-making. Moreover, it underscores the relationship between discretion and well-being, which has broader implications for the public service workforce.
Economic Security and the Birth of a Child: Three Essays on Employment, Income, and Paid Leave Among Parents of Newborns
(2024-09-09) Pelletier, Elizabeth Froom; Allard, Scott W.
This dissertation focuses on parents of newborns around the time a child is born, examining the economic conditions of parents around a birth and exploring the potential of public policies to promote economic stability for families around this time. The first chapter uses administrative microdata from birth certificate and earnings records to analyze patterns of earnings and employment among parents who welcomed a new baby between 2010 and 2016 in Washington State. I present a detailed analysis of parental earnings patterns around childbirth, tracking employment rate, earnings and hours levels, employer continuity, and earnings volatility in the two-year period around a birth for all parents and for mothers disaggregated by racial and ethnic identity, educational attainment, and wage rate quintile. Next, I examine household-level measures of earnings and safety net program income in the time around a birth. I find that parents in Washington State experience substantial earnings volatility when a child is born. Aggregating earnings of both parents listed on a given birth certificate to calculate household-level income measures does attenuate the volatility faced by individual parents, but substantial economic instability still remains. For example, over half of Washington households see their income fall by half or more during the year around a birth. Use of means-tested programs increases in the perinatal period, which offsets earnings volatility slightly (especially for single mother households) – but household earnings remain volatile even with the addition of these sources of income. The second chapter studies paid leave programs, which many states have recently enacted. Paid leave policies differ significantly across states, and the implications of these policy differences remain under-studied. This paper assesses the implications of state policy design features for the share of parents eligible for paid leave, as well as for disparities in eligibility across parent demographic, socioeconomic, and employment characteristics. I use administrative microdata from Washington State to simulate working parents’ eligibility for claiming paid leave under ten different states’ policies. State policy designs differ dramatically in terms of overall generosity and disparities in eligibility across subgroups. In general, policies that are less restrictive and allow more parents to access paid leave also significantly narrow disparities in who is eligible. State policies with more stringent employment requirements disproportionately exclude mothers (versus fathers); mothers working in low-wage jobs; mothers with lower educational attainment; and Black, Indigenous, and Latina mothers from qualifying for paid leave. Finally, the third chapter focuses on one such paid leave policy, Washington State’s Paid Family and Medical Leave (PFML) program. Starting in 2020, parents who worked at least 820 hours in the year before a birth qualify for up to twelve weeks of paid leave to bond with a new child, and mothers can take additional leave for pregnancy and related health conditions. I describe use of the policy in its first few years, using multiple sources of administrative microdata: health insurance data on birth-related insurance claims from the Washington State All Payer Claims Database and employment and paid leave records from the state’s Employment Security Department. These records enable me to identify a population of potential policy users, precisely estimate policy eligibility and take-up, and estimate how PFML use affected employment. I find that a majority of eligible Washington mothers who gave birth between 2020 and 2022 use PFML at some point during the perinatal period. Take-up increases between 2020 and 2022 as the policy rolls out, especially for medical leave. Analysis of policy take-up by mother characteristics reveals some important disparities. For example, results suggest that eligible Native Hawaiian/Pacific Islander and American Indian/Alaska Native mothers are less likely to take up paid bonding and medical leave than mothers identifying with other racial and ethnic groups. Mothers in more urban areas are more likely to take up both types of leave. There are also sharp disparities by wage rate; for example, only 30 percent of eligible mothers in the lowest-wage jobs took up bonding leave compared to 67 percent of mothers in the highest-wage jobs. Regression discontinuity analyses of the local average treatment effect of PFML eligibility on employment among mothers around the eligibility threshold find mixed results. There is some evidence that eligibility for PFML has small negative effects on employment status and intensity (i.e., hours worked) in the short term. However, this largely does not translate into significant reductions in total earnings (including wages plus PFML benefits), suggesting that mothers who are just eligible for the policy, on average, are able to spend more paid time off with their children without seeing reductions in overall income. Among mothers who worked following a birth, PFML eligibility led to an increase in continuity of work with the same employer.
Essays on Government-led Collaborative Governance
(2024-02-12) Kim, Yulan; Herranz Jr., JoaquÃn
Governments increasingly use collaborative governance as a strategy for policymaking and management. This dissertation is a collection of three essays each addressing an important question regarding government-led collaborative governance. Social Security Consultative Bodies (SSCBs) in South Korea—a case of government-mandated collaborative governance—constitute the empirical context of the essays. The first essay explores whether governments serving as collaborative platforms can breed trust using repeated measures survey data. Findings show that as a result of participating in government-led collaborative governance, SSCB participants report an increase in both institutional and individual-level trust. The study confirms that governments serving as collaborative platforms can initiate genuine collaboration as evidenced by enhanced trust and identifies factors that government platforms can leverage to facilitate trust building at different levels. The second essay explores the policy feedback effects of government-led collaborative governance. The central question is whether government-created venues crowd out or reinforce participation in alternative participatory venues. A mixed-methods analysis using both survey and interview data uncovers that government-led collaborative governance generates both resource effects and interpretive effects which promotes further civic engagement. The findings imply that government-created collaborative venues generate cooperative feedback effects that reinforce other participatory venues. The final essay focuses on the co-production function of SSCBs. I explore what implications service delivery through government-led collaborative governance has for place-based inequities in access to public services. Using survey data, I find that internal process dynamics are associated with co-production performance in delivering public services, rather than local resource capacities. This implies successful management of government-led collaborative governance can mitigate place-based inequities that arise due to local resource disparities. Collectively, the essays advance our knowledge of a particular type of collaborative governance that is government-led while making both theoretical and methodological contributions to the study of public management.
Multidimensional Approaches to Nonprofit Revenue-Generation Strategies and Outcomes: Organization, Community, and Institution
(2024-02-12) Park, Gowun; SuÃ¡rez, David F.
Among all types of resources, revenue stands out as one of the most crucial assets for nonprofit organizations to fulfill their distinctive missions, sustain their operations, and ensure their survival and long-term viability. Given the pivotal role of revenue in lifecycle of organizations, how do nonprofits enhance their performance in acquiring it? What factors influence the outcomes of revenue generation for nonprofits? This dissertation investigates how various attributes constituting nonprofits’ resource environment – including organizational, community, institutional characteristics - influence revenues strategies and outcomes, employing a range of theoretical frameworks and empirical evidence. Firstly, this study examines the effects of management characteristics on the extent of nonprofit revenue diversification, focusing on nonprofits’ managerialism, collaboration, and community ties. This study theoretically and empirically supports the idea that nonprofits utilize managerialism and collaboration as strategic tools to generate diverse revenue streams in their unique multilevel environmental contexts. Secondly, this dissertation explores how these multilevel factors affect the likelihood of human service nonprofits securing government contracts. It operates under the theoretical assumption that government cost considerations, nonprofits’ resource factors, and institutional pressures play key roles in determining the success of nonprofits in securing public contracts. The research demonstrates that nonprofits’ utilization of performance measures, administrative capacity, and advocacy efforts increase their chances of securing public contracts by aligning with the requirements of multiple levels. Thirdly, this study delves into the influence of social capital at the community level on charitable giving and foundation grants for nonprofits. The research findings suggest that nonprofits situated in communities with high rates of volunteering are likely to have a larger portion of foundation funding, while a high density of nonprofits has a negative influence. Conversely, tightly knit communities with dense social clusters tend to offer fewer contributions of individual donors. Additionally, the paper examines how leadership diversity connects social capital to nonprofits’ financial performance. It finds that board diversity significantly moderates the impact of bridging on individual donation income and the impact of bridging and bonding on foundation grant income. In summary, this dissertation provides valuable insights into the intricate web of relationships nonprofit organizational attributes, community characteristics, institutional environment, and nonprofits’ revenue acquisition outcomes. It underscores the adaptive nature of nonprofits in response to their environments and emphasizes the significance of strategic decision-making in ensuring their long-term sustainability.
Three Essays on Income Support Policy, Family Violence, and Household Economic Arrangements
(2022-09-23) Kovski, Nicole Lynn; Klawitter, Marieka M
This dissertation focuses on the relationships between income support policy, economic security, and family dynamics that contribute to health and wellbeing. Chapters 1 and 2 examine potential consequences of exposure to tax credits, the Earned Income Tax Credit (EITC) and the Child Tax Credit (CTC), for child maltreatment reports and intimate partner violence – two outcomes that are strikingly common in the United States and disproportionately experienced by those in low-income households. The third chapter is also related to the economic security of families. Typically, in studies of economic-wellbeing, measurement tends to occur at the household level, thereby obscuring patterns of ownership of economic resources within families. In chapter 3, I examine trends and patterns of liquid asset ownership within couples with children, with particular attention to how these patterns are shaped by gendered power dynamics. Chapter 1 leverages a natural experiment – created by a legislated change in the timing of the annual disbursement of EITC and CTC transfer payments to families – to estimate the association between EITC and CTC payments and child maltreatment reports in the period shortly after families receive payments from these programs. Using weekly tax refund data from the Internal Revenue Service linked to state-specific child maltreatment report data from the National Child Abuse and Neglect Data System during the 2015 through 2018 tax seasons, I find evidence that reports of child maltreatment made to child welfare authorities decline in the weeks immediately following issuance of these lump-sum credits. My results imply that child maltreatment risk is responsive to not only chronic economic hardship but also to more immediate income availability among caregivers. Chapter 2 focuses on pregnant women’s exposure to intimate partner violence, which affects maternal health and birth outcomes. Using individual-level data from the 1996-2018 Pregnancy Risk Assessment Monitoring System (PRAMS) and a difference-in-differences methodological approach, I find that expansions to the EITC were associated with lower risk of self-reported physical abuse during pregnancy. These effects are concentrated among mothers who are most likely to receive EITC benefits during pregnancy – those with lower levels of education and previous children. Given a broader literature documenting improvements in maternal-infant health associated with EITC expansions, reductions in violence against pregnant women may be one channel through which the EITC leads to maternal-infant health improvements. Chapter 3 uses data from the 1998-2019 waves of the Survey of Consumer Finances to examine patterns and trends in asset ownership within couples with children. Drawing on theories of bargaining power, I test the premise that differentials between partners – in their education, employment, and health status – are associated with each partner’s liquid asset ownership. I find that the proportion of couples with children who hold liquid assets in separately owned accounts has increased between 1998 and 2019. Both men and women have increasingly used separately owned accounts, but the median account balance was greater for men than women. I also find that proxies for intra-partner bargaining power are associated with the likelihood that women, but not men, separately own liquid assets. This study demonstrates the importance of considering intrahousehold allocations in studies of economic wellbeing. Together, these chapters contribute to our understanding of the downstream effects of income support policies for the health and wellbeing of recipients and their children as well as demonstrate the importance of considering how intrahousehold dynamics might condition the effects of such transfers.
The Color of Homelessness: The Causes, Reproduction, and Consequences of Racial Inequality in Homelessness
(2022-09-23) Fowle, Matthew Z; Fyall, Rachel
People of color or mixed race account for more than half of all people experiencing homelessness, yet comprise less than a quarter of the total population in the United States. Despite these massive racial disparities, there is a lack of research examining the intersection between race and homelessness. In this dissertation, I conduct three studies that each answer a crucial question regarding racial inequality in homelessness: 1) what causes it? 2) what reproduces it? and 3) what are its consequences? In the first study, I conduct a literature review of historical and contemporary research to show that three primary systems of stratification drive racial disparities in homelessness: racial economic inequality, housing discrimination and residential segregation, and the homeless response system. In doing so, I recast homelessness as a deeply racialized form of inequality that has existed at least since White settlers colonized North America and established the institution of slavery. In the second study, I draw on an original survey (n=410), administrative data (n=7,140), and in-depth interviews (n=25) to reveal that low-income tenants faced a 179% increase in the odds of experiencing an informal eviction tactic during the first year of the COVID-19 pandemic. Black and Hispanic/Latinx movers were also significantly more likely to experience a forced move than White movers during the pandemic. In the third study, I construct a novel dataset of homeless deaths across 23 major U.S. counties (n=16,874) to demonstrate that people experiencing homelessness are, on average, 3.1-7.2 times more likely to die than the general population. Among people of color, “natural” deaths from preventable conditions such as cardiovascular disease, respiratory illness, and exposure to environmental conditions are significant contributors to this rise in mortality, in addition to “deaths of despair.” I estimate a total of 276 excess deaths during the COVID-19 pandemic, only 15.6% of which were due to COVID-19. Together, these studies provide novel insight into the extent to which racial inequality in homelessness is an outcome of inequities produced by social institutions and public policies, a factor in reproducing downward intergenerational mobility, and a driver of premature death among households of color.
Three Essays on Preferences for and Determinants of Participatory Philanthropy in U.S. Foundations
(2022-09-23) Finchum-Mason, Emily; Gugerty, Mary Kay; Suarez, David F.
Philanthropic foundations are an important, if understudied, governance institution in the United States that exert tremendous influence over public policy and political discourse through grant-making, capacity-building and technical support, and by shaping organizational fields. Historically, these institutions have been closed off to public input, instead relying on professional expertise to dictate policy problems and their appropriate solutions. But movements to open governing institutions and make them more accessible to citizen/beneficiary/client input have fomented over the past decades. The nature of public challenges has also become increasingly complex and boundary-spanning, necessitating more collaborative models of governance that engage all parties, including those who have been historically marginalized. There is a wealth of literature on how these movements for open governance and citizen participation affect the public sector and there is even growing pracademic literature about participation in grant-making public charities and nonprofits. There is, however, very little research on how participation manifests in the context of private philanthropy. This dissertation seeks to fill this gap by exploring how foundations value stakeholder participation and the organizational factors that predict utilization of participatory practices in these large foundations. Outcome data on participatory practices used to satisfy these research objectives comes from a novel survey of the 500 largest private and community foundations in the United States (by total assets, 2018). Conducted in 2020, this survey represents the first systematic attempt to study the participatory practices of large foundations in the United States. Explanatory variable data on foundations leaders, staff, and organizational attributes was obtained using foundation websites, LinkedIn, and social media aggregators. Overall, the dataset used in this dissertation provides insights on foundations’ leaders, staff, organizational structure, and decision-making processes – this is a unique contribution to a field that has been understudied primarily because of the inherent difficulties of obtaining information about these intensely private organizations. In aggregate, findings from this series of studies build upon previous research and suggest that participatory philanthropy is not only a part of many foundations’ governance repertoires, but as Chapter 2 illustrates, that foundations clearly value stakeholder participation in their grantees. The limitation is that this support decreases as a function of the amount of power-sharing between the grantees and those they serve. That is, foundations express more support for the idea of their grantees consulting beneficiaries than devolving decision-making power to those beneficiaries. That foundations are demonstrably value stakeholder participation, suggests a potential willingness to experiment with and learn from stakeholder participation and to cede power over internal decision-making processes to some extent. Chapters 3 explores one mechanism theorized to drive the adoption and utilization of stakeholder participation in U.S. foundations – namely, the association between foundation program staff characteristics and the uptake/utilization of participatory practices in governance and grant-making. Program staff are the frontline workers of the foundation, essentially the foundation’s equivalent of the street-level bureaucrat; they cultivate relationships with grantees and manage the disconnect between the way that foundation leadership conceptualizes social problems and the way these problems are understood by those who face them. Three attributes – the proportion of program staff of color, the proportion of program staff educated at elite institutions, and the proportion of program staff educated in business – are tested to determine whether and how they influence organizational-level participatory philanthropic outcomes. Program staff race is found not to be associated with whether a foundation utilizes participatory practices but is significantly positively associated with the extent to which foundations use these practices, a finding which merits further exploration, particularly given the challenges that people of color can encounter working in predominantly white, patriarchal institutions (see Villanueva, 2018 and Kohl-Arenas, 2017). The proportion of program staff educated at elite institutions is found to be negatively associated with uptake and utilization, such that a greater proportion of Ivy League staff is associated with a smaller likelihood and extent of utilization of participatory practices. Finally, the proportion of program staff who are socialized in the business discipline, which is expected to be negatively associated, is positively associated with the extent of utilization of participatory methods. These findings provide some preliminary evidence that diverse worldviews of program staff are indeed correlated with stakeholder participation efforts within the foundation, but more work remains to ascertain whether these relationships are causal. Chapter 4 examines whether and how female leadership and female leaders of color are associated with uptake and utilization of participatory philanthropic practices. Feminist approaches to philanthropy and participatory philanthropy dovetail in terms of their egalitarian, democratic approaches to giving. In fact, both are predicated on the logic of sharing power between resource holders and the historically marginalized. There are also notable observable differences in female leadership behaviors, namely the observed propensity to demonstrate more transformative leadership characteristics, valuing relationship-building and subordinate empowerment more than male leaders in comparable positions. This study theorizes that foundations headed by female executive directors (and female directors of color) may be more likely to utilize participatory approaches to philanthropy. An important counterargument, however, is that organizational inertia and the barriers that women face in terms of influence may be too great to overcome. Indeed, this study finds female leadership and female leadership of color to be unassociated with either the adoption or extent of utilization of participatory philanthropic methods at any level of power-sharing with stakeholders. Rather than seeing these findings as an indication of a summary lack of impact of female leaders, this work raises questions about the organizational factors that impede or facilitate women’s influence and invites further exploration into the leadership-level determinants of utilizing democratizing philanthropic practices.
Essays on Food Insecurity and Food and Nutrition Assistance Policy in the United States
(2022-01-26) Charnes, Sarah Elizabeth; Vigdor, Jacob L
This dissertation investigates many facets of means-testing in the United States through the lens of public food assistance. In Chapter 1, I speak to the literature on “administrative burden,” or individual-level barriers to means-tested program participation. Previous studies debate the extent to which administrative barriers inhibit take-up of means-tested programs. I study two application streamlining initiatives intended to simplify the Supplemental Nutrition Assistance Program (SNAP) application process in the United States through the reduction of transaction and information costs. The two initiatives differ along the dimension of in-person versus mail-based interactions with clients. Using two-way fixed effects and alternative difference-in-difference estimators, I estimate an overall 4.3 percentage point (19.3 percent) average treatment effect of application streamlining on SNAP participation. Further analysis of the two implementation models suggests a stronger effect of in-person interactions with clients (25.8 percent), compared to off-site outreach (15.2 percent). However, different approaches appear to be more effective for different eligible populations: there is suggestive evidence that off-site outreach could have a stronger effect for population subgroups experiencing mobility-related barriers to take-up. As such, this study points to the importance of understanding the behaviors and barriers to take-up experienced by specific target populations when designing initiatives intended to improve enrollment in means-tested programs. In Chapter 2, I speak to current discourse around the association between household food insecurity and disability status. Disability is a known risk factor for food insecurity, even when accounting for household income. However, the mechanisms driving the relationship between disability and food insecurity remain underexplored. Using the National Household Food Acquisition and Purchase Survey, I test the extent to which food store choice (representing food access) mediates the association between disability and food insecurity in the United States. The analysis is complicated by the notion that food insecurity also influences food store choice. Nevertheless, multivariate regression findings suggest that food access is not a significant driver of high rates of food insecurity among households where disabilities are present. This chapter has been accepted for publication in Physiology & Behavior (Charnes, forthcoming). In Chapter 3, I address questions surrounding the cause of the SNAP benefit cycle – a phenomenon in which SNAP benefits (disbursed on a monthly basis) are typically spent all at once within the first few days of receipt. The disbursement of Supplemental Nutrition Assistance Program (SNAP) benefits is associated with a decline in food spending and caloric consumption over the SNAP month, resulting in a range of adverse consequences. However, there is a lack of consensus about the underlying cause of the SNAP benefit cycle. Building upon work conducted by Tiehen, Newman, and Kirlin (2017), I use the National Household Food Acquisition and Purchase Survey to examine SNAP households’ acquisitions of free food patterns across the SNAP month. I conclude that a steady state of free food acquisitions across the month is primarily attributable to benefit inadequacy. Although the three chapters are situated within distinct sets of literature, they jointly point to the importance of public food assistance for Americans in need. This dissertation was written during the Trump Presidency, which was characterized by movements to drastically cut the social safety net – followed by the COVID-19 pandemic, its associated recession, and movements to rebuild the safety net in the early years of the Biden Presidency. The three essays highlight the conditions that have led to current proposals to transition to a universal structure for SNAP and other safety net programs.
Households' perceptions of earthquake risks and protective measure adoption intentions
(2022-01-26) Ahn, Young Eun; Bostrom, Ann
Preparedness levels vary among individuals who face common threats of natural hazards in the region in which they live. The studies in this dissertation explore the factors that may explain such differences, examining the relationships between key predictors of protective action decisions and protective behavior intentions in the context of Western Washington. Specifically, this dissertation aims to answer the following research question: How do Western Washington residents understand earthquake risks and form intentions to adopt protective measures? Chapter 2 investigates the effect of perceived efficacy of seismic reinforcement on people’s intentions to pay for such protective measures using the Meta-Cognitive Model (MCM) approach. Using original data collected in King County, WA, the study examines the association between perceived efficacy and intended payments for seismic retrofits jointly with one’s metacognitive confidence in their assessment of efficacy using interaction effects. The hypothesized moderating effect of metacognitive confidence is not confirmed, but perceived efficacy is found to be an important determinant of intentions to pay for retrofits. Chapter 3 provides an international comparison of perceptions of earthquake risk, efficacy of earthquake early warning (EEW), and willingness to pay (WTP) for EEW between the city of Sendai, Japan and Seattle, WA that have similarities and differences in terms of earthquake hazards and availability of EEW. Risk perceptions and perceived effectiveness of EEW in personal protection are found to be significant determinants of WTP for EEW in both cities, while the association of fear with earthquakes is a significant predictor of WTP only in Seattle. In addition, this study uses a double hurdle model to examine if people’s WTP is a two-stage decision that involves (a) an assessment of the determinants of the decision to pay nothing versus anything at all, and (b) among those who decide to pay something, an assessment of the factors that are associated with how much one is willing to pay. The study finds that the set of variables predicting whether one decides to pay or not does not predict the amount of payment. This suggests that indeed, two sets of decisions may be involved. Finally, Chapter 4 uses a choice experiment to investigate (a) how the key predictors of protective action affect the probability of one choosing to adopt a resource-intensive protection measure over taking no action, and (b) whether residents of Western Washington are willing to pay for resource-intensive measures. Perceived efficacy of each protective measure is found to be positively associated with the probability of one choosing the respective measure over taking no protective action. However, residents of King County are generally not willing to pay for resource-intensive measures, suggesting a need for policy to increase uptake of protective action in the region.
Essays on Women’s Control over Income, Maternal Cash Transfers, and Rainfall Shocks: Evidence from India and Malawi
(2021-10-29) Patwardhan, Vedavati; Anderson, C. Leigh; Gugerty, Mary Kay
In this dissertation, I conduct three studies on women’s economic empowerment in India and Malawi. The first study examines the effect of an Indian maternal cash transfer program on child nutrition. The second study explores variation in women’s control over income by income type and family structure in Malawian households. Also in Malawi, the third study analyzes the effect of rainfall deficits on women’s control over income. In my first chapter, I ask: 1) What are the effects of a universal maternal cash transfer program “Mamata Scheme” in Odisha, India on child nutrition? 2) To what extent do program effects vary by household wealth? I use a triple differences estimation and nationally representative survey data from the National Family Health Survey to test the effect of the Mamata Scheme on standardized height and weight measures for children under five. I find improvements in some, but not all, measures of child nutrition following the implementation of the Mamata Scheme. I find evidence of wealth heterogeneities in the program’s effects. Improvements in child nutrition are concentrated in children from non-poor households. Taken together, the results suggest that maternal cash benefits improve child nutrition, but universal schemes such as Mamata may need to offer additional incentives and/or improve targeting for children from poor households to realize program benefits. In the second chapter, I ask: 1) Does women’s control over household income differ by income type? 2) Is there a relationship between family structure, women’s demographic characteristics, and women’s control over income? I use the Fourth Malawi Integrated Household Survey (IHS-4) to examine women’s control over various sources of household income, including public and private transfers, farm, and non-farm earnings. Using descriptive and multivariate logistic regression analyses, I find that in households with both adult men and women, women have higher odds of having sole control over transfer income, compared to income from farm and non-farm sources. I also find that the presence of a male spouse (versus another adult man) in the household is associated with significantly lower sole female control over transfers. Further, I find that female characteristics, especially age, divorce, and widowhood increase the probability of controlling transfers. The results shed important light on women’s financial decision-making patterns in agricultural households. Specifically, household income received through remittances and transfer programs is associated with higher female decision-making authority, a key insight for policy interventions aimed at increasing women’s control over financial resources. In the third chapter I ask: 1) How do negative rainfall shocks affect women’s control over household income? 2) Do effects vary by income type? Using three rounds of the Malawi Integrated Household Panel Survey (IHPS), I employ a household-year fixed-effects identification to test whether female control over farm, nonfarm, and transfer income changes in response to drought shocks. In dual-adult households, a drought shock increases women’s control over farm income, while there is no change in female control over total household income. I also find that rainfall shocks increase the probability of a household having a farm plot that is solely managed by a woman. The results suggest that drought alters intra-household decision-making, particularly related to the farm. While control over income is typically used as a proxy for women’s economic empowerment, the findings suggest that changes in control over income in response to climatic factors such as rainfall variability may not reflect an absolute improvement in women’s empowerment.
The Impact of Public Policy on Nonstandard Work Arrangements
(2021-10-29) Glasner, Benjamin; Long, Mark
The social compact between workers, firms, and governing bodies has been typified by the balancing of market efficiency, social fairness, and bargaining power. This balancing act has produced the minimum wage, overtime pay, protections from discrimination in hiring and firing, unemployment insurance, social security, and so much more. Yet, whether it is through an explicit terminology of employment or an implicit organizational structure, the classification of who is an employee remains a critical component of who has access to the benefits of this bargaining effort. The criteria of this classification has changed over time with expansions in the Fair Labor Standards Act, the common law agency test, the economic reality test and the ABC test (Shimabukuro, 2021). This dissertation explores the ways in which public policy impacts the less-conventional forms of work in the United States. The chapters of this dissertation draw on prior work relating to the minimum wage, employer supplied health insurance, nonstandard work arrangements, monopsonistic competition, and tax evasion to examine the impacts of public programs and government regulation of the labor market on self-employment and multiple jobholding.
Foundations for Change? Advocacy and Participatory Grantmaking Strategies in the U.S. Philanthropic Sector
(2021-10-29) Husted, Kelly; Suarez, David F
Philanthropic foundations are not merely patrons that provide grant funding to worthy causes. Iseek to develop an understanding of philanthropic foundations as active political actors that attempt to influence social and policy change through two key strategies: advocacy and participatory grantmaking. The role of foundations in social and policy change merits scholarly attention for several reasons. For one, foundations wield significant power with the nearly $1 trillion in assets that they collectively hold, and they can wield this power to not only define the nature of societal challenges but also the manner in which they are addressed. Foundations—in particular private foundations—are also a unique organizational type as they are typically seeded with money from the wealthy, who receive generous tax breaks for doing so, and then are often run by the wealthy with minimal oversight. Critics argue that they thus represent a plutocratic and undemocratic force in U.S. society. Furthermore, their role as funders aims the attention of the nonprofit sector as a whole to focus on particular challenges and solutions and not on others. I develop a conceptual framework in the first chapter that aims to further understandingof what explains foundations engagement in social and policy change strategies. Drawing primarily on theories of organizational sociology, the framework posits that foundations’ structure and context influence their engagement in advocacy and participatory grantmaking. With respect to structure, I expect that foundations’ institutional logic, organizational form, levels of managerialism, and political orientation will affect their social and policy change strategies. In regard to foundations’ context, I expect that relational factors—network embeddedness and organizational field—as well as external environment factors—community wealth and community political orientation—will influence foundation engagement in this work. I test the framework across three empirical studies. Chapter two examines the extent to which the structure and context of large foundations influences their engagement in advocacy. Chapter three narrows the focus to community foundations, specifically, and their engagement in advocacy and lobbying. Chapter four turns back to large foundations to examine their adoption of participatory grantmaking. Across the findings from the three empirical chapters, the conceptual frame appliessomewhat differently among these contexts, but there are key themes across the studies. A community logic is consequential for advocacy as we see for large foundations broadly and community foundations specifically. Foundations with a community logic, and its focus on partnership and collaboration, engage in more advocacy than other foundations. Embeddedness in inter-organizational networks is also an important driver of advocacy among both large foundations and community foundations, lending support to the idea that networks are important conduits for knowledge sharing about advocacy. Furthermore, foundation form matters for both advocacy and participatory grantmaking but in divergent ways. Community foundations engage in less advocacy than private foundations but are more likely to embrace participatory grantmaking. In this regard, community foundations may feel pressure from donors and stakeholders to refrain overtly political activity, whereas participatory grantmaking, even if used as a tool for social change, may seem more in line with the roles and values of community foundations. The political or social justice orientation of foundations also clearly matters for bothadvocacy and participatory grantmaking. Across the three chapters, foundations with a liberal or social justice orientation are more likely to display social and policy change discourse. However, with respect to concrete actions such as advocacy funding, lobbying, and participatory grantmaking, they do not engage in these strategies more than other foundations. Conversely, conservative foundations are not associated with advocacy discourse but they are strongly associated with advocacy funding. While discourse is certainly important as it can reflect the values and actions of foundations, these findings across multiple studies should promote reflection among liberal foundations about whether their actions for social and policy change match their public presentation. The dissertation makes several contributions to the field. First, in terms of theoreticalcontributions, my dissertation applies theories from organizational sociology to this unique—yet understudied—type of organization. The dissertation also advances the nonprofit advocacy literature to address the drivers of foundation advocacy. Until now, this research has focused primarily on the determinants of advocacy among service-providing nonprofits. From a methodological standpoint, the dissertation uses machine-learning techniques to develop a replicable measure of advocacy that can be used for any organization with a website, and it generates original survey data on this secretive and hard-to-reach population. From a practical perspective, the dissertation raises both practical considerations and normative concerns for foundations engaging in these strategies.
Pueblos Bicicleteros: Three Essays on Cycling Policy in Mexican Cities
(2021-10-29) de Buen Kalman, Rebeca Cecilia; Cullen, Alison C
In this dissertation, I investigate cycling policy adoption, design, and implementation in Mexico. In Chapter 2, I research the interactions between driving restrictions and bikeshare usage in Mexico City. The Mexico City government has introduced policies to reduce pollution from cars, including a license-plate based driving restriction and a bikeshare system. When restrictions are in place, people need to find transportation alternatives. In this research, I leverage the random nature of driving restrictions to explore whether people use bikeshare differently when driving is restricted due to poor air quality. I use negative binomial models to study these effects using data from 2016-2019. Results indicate that restricted days exhibit approximately 17% lower bikeshare use (10 - 24% decrease, 95% CI) for non-peak traffic hours compared to regular days. However, ridership increases during peak traffic hours. Morning ridership increases up to 12.5% (3 – 22% increase, 95% CI) and evening ridership increases up to 16.2% (5 – 27.6% increase, 95% CI). The analysis suggests that the poor environmental conditions may buffer bikeshare system use increases and that bikeshare can be a critical partner in local transportation infrastructure.In Chapter 3, I researched cycling infrastructure implementation in ten cities in Mexico using a multiple case study research design and thematic analysis. The questions motivating this research were: What is the role of civil society organizations in the process of infrastructure delivery? Why and how do governments implement cycling infrastructure? In the first part of this chapter, I analyzed the role CSOs have played in developing cycling infrastructure in Mexican Cities. While the presence or activity of civil society organizations did not guarantee the implementation of cycling infrastructure, this research demonstrated that in most settings, CSOs are not only actively involved in every aspect of infrastructure provision and its institutionalization as a governmental activity but represent an essential presence in ensuring progress and the professionalization of infrastructure design. In the second part of chapter 3, I studied how infrastructure is implemented in each city included in this study: Cuernavaca, Toluca, Oaxaca, MÃ©rida, Aguascalientes, QuerÃ©taro, Morelia, LeÃ³n, Puebla, and Guadalajara. I studied the primary laws, organizations, and planning instruments and norms that cities have used to implement cycling infrastructure and policy successfully. I found that high-level mandates in state laws on their own make very little difference in terms of making progress on the ground both for kilometers of infrastructure implemented and its quality. Lack of laws can be a barrier, for example, for garnering funds to build infrastructure, limiting or slowing down the ability to implement projects, but this is not always the case. Specialized agencies (Municipal Mobility Offices and State Mobility Agencies) containing non-motorized mobility departments have proven to be one of the most important variables promoting the implementation of cycling infrastructure. The agencies responsible for designing and implementing projects are critical for sustaining cycling infrastructure planning and implementation. Finally, in Chapter 4, I studied the institutionalization of cycling policy in Guadalajara. Over the past 20 years, Guadalajara Jalisco has gone from being a city with few cyclists and no public policy or funding to support cycling as a transportation mode to a nationally and internationally recognized city for its work in advancing cycling mobility. The research questions guiding this case study are: what were the main factors and events that have led Guadalajara to adopt and implement policies to promote cycling mobility? What is the story behind Guadalajara’s adoption and implementation of cycling policy? What evidence exists about the success of these policies? In this paper, I analyzed the process that led to this urban and institutional transformation of Guadalajara. This analysis identifies the actors and events that have made Guadalajara a reference as a cycling city and presents salient evidence on the impact of these programs. I demonstrate how local actors built a strong, diverse, and highly media driven movement to mobilize sustainable transportation as a new policy issue and successfully pushed for its gradual institutionalization as a core part of Guadalajara’s metropolitan urban planning agenda.
Criminal Justice Policy, Race, and the Dynamics of Admission to Prison
(2020-10-26) Obara, Emmi; Allard, Scott
Research has long documented racial and economic disparities in imprisonment and the recidivism rate. Despite studies that suggest these disparities cannot be fully explained by differential involvement in crime, scholars have not paid enough attention to how individual characteristics might interact with macro-level factors. This dissertation adds to the literature by examining interactions between individual-level characteristics such as race and history of prior criminal justice contact, and macro-level factors such as criminal justice policies and local racial composition. I structure this dissertation with three chapters that are increasingly focused, in terms of policy and geography. Chapter One extends criminological research investigating the relationship between racial composition and phenomena in the criminal justice sphere. This chapter investigates how county racial composition affects racial disparities in recidivism. The analyses test two competing explanations: racial threat, which would predict a positive relationship between the relative size of the nonwhite population and racial disparities in returns to prison, and political representation, which would predict a negative relationship. I use National Corrections Reporting Program (NCRP) data from 18 states and event history methods to test my hypotheses. The results show that increases in the percent of a county’s Hispanic population are associated with a small increase in the hazard of recidivism for Hispanics and Blacks compared to Whites, which is consistent with the racial threat hypothesis. The results also show that an increase in the non-Hispanic Black population is associated with a small overall decrease in the hazard of recidivism as well as a small decrease for non-Hispanic Blacks compared to Whites, which is consistent with the political representation explanation. Future research including other explanatory variables around political representation would further this study. For example, examining the effect of the racial composition of county-level elected officials or the proportion of adults barred from voting due to a jurisdiction’s felony disenfranchisement laws would be helpful. The second chapter focuses on parole, a period of conditional supervised release in the community following a prison term. Returns to prison while on parole are called revocations, and this can occur in two ways. A parole revocation can result from a new crime conviction, or from a technical violation. Technical violations occur from failing to adhere to conditions that typically would not incur a prison term, such as not completing substance abuse treatment or possessing a firearm. Parole revocation policies vary highly across jurisdictions and include differences regarding the importance of past parole terms (episodes). This chapter advances scholarly understanding of how state differences in parole policy and implementation contribute to differences in the likelihood of experiencing a return to prison. Analyses focus on how the number of parole episodes a person has served interacts with the state they are in to affect the likelihood of experiencing a parole revocation. Comparing jurisdictions in this way can reveal how institutional context and differences in policy shape the outcomes of individuals. It further contributes to the literature by distinguishing between revocations for new crimes and for technical violations, which most studies do not do. Unique parole data for New York and Pennsylvania in the NCRP provide the opportunity to examine revocations in two different institutional contexts. Findings suggest the number of prior parole episodes and the state is significantly associated with differential likelihood and timing of returning to prison and revocation type. Chapter Three focuses further to understand variation within one state, by exploring drug courts in Georgia. Drug court diversion programs offer people charged with drug possession who have a substance abuse issue an alternative to prison in exchange for participating in a court-supervised intensive treatment program. However, these programs come with high fees that participants must shoulder. If people who do not have the ability to pay the participation fees are excluded from accessing this process of remaining out of prison, drug court implementation may lead to an overall increase in the level of economic disadvantage among people sent to prison for drug possession. Thus, while existing research on drug courts focuses on whether they reduce recidivism for participants, I empirically test whether drug courts increase economic inequity by considering who is excluded from participation. I use a difference-in-differences strategy exploiting time variation in the implementation of drug courts across the state to answer this question. I find that on average, the level of disadvantage as measured by educational attainment does not increase significantly among people admitted to prison for drug possession due to drug court implementation. Taken together, the findings of this dissertation suggest policy makers and advocates should consider policy levers that address structural factors and not focus solely on programs aimed at altering individual behavior. Scholars should further examine how macro-level factors contribute to recidivism as well as their interactions with individual characteristics to better understand how we can reduce racial and economic disparities in the prison system as well as the cumulative effect of cycling through prison.
Too Much Risk? Public Policy, Employment, and Earnings Volatility in Washington State
(2020-08-14) Wething, Hilary C; Hill, Heather D
In this dissertation, I conducted three studies of the causes and consequences of earnings volatility for workers in low-wage jobs using of administrative records from Washington State’s Unemployment Insurance (UI) program between 2006 and 2016. In my first study, I asked: 1) What were the trends in intra-year earnings volatility for workers in low-wage jobs in Washington State between 2006 and 2016? 2) I used descriptive analysis and subsequently decomposed workers’ intra-year earnings volatility to estimate the contribution of between-work volatility, the volatility that occurs from transitioning in and out of the UI-covered jobs, and within-work volatility, the volatility that occurs from fluctuations in hours and wages within continuous UI-covered employment. Among workers who were continuously employed during that period, I further estimated the contribution of workers’ wages and hours volatility to their earnings volatility. I found that earnings swings (an earnings change > 25 percent of the previous quarter's earnings) were frequent for workers in low-wage work. Workers were more likely to experience large declines during the economic downturn (2007-2009) and large increases during the period of strong labor market growth (2012-2016). However, the magnitude of large increases was smaller than those of large decreases. I further found that worker transitions in and out of UI-covered work—15.6 percent of all quarter-to-quarter transitions—accounted for nearly two-thirds of the total intra-year earnings volatility documented over the period. Among the subset of workers who experienced continuous employment, I found that volatility from workers' hours worked had a significant, positive association with workers’ intra-year earnings volatility. In the second chapter, I asked 1) What was the impact of a local paid sick leave policy on the employment levels, flows, and volatility of affected firms and workers? 2) Were these impacts larger for workers less likely to have access to paid sick leave before the policy? My analytic strategy was designed to evaluate changes in employment outcomes for firms and workers affected by the mandate (firms with at least four FTE employees) relative to those not covered by the policy (firms with four or fewer FTE employees) between 2010 and 2014. I focused on traditional measures of employment flows, such as hires, separations, and job duration, and on new measures of employment volatility, such as the quarterly arc percent change in workers’ hours and earnings and their likelihood of experiencing a drop or gain in these outcomes. I found that firms right above the threshold of four full-time equivalent employees did not experience any statistically significant changes in their total headcount, payroll, hires, separations, or job turnover resulting from the policy. Workers in firms just above the FTE employment threshold for the PSST mandate experienced a 0.1 percent decrease in their hours worked in the year after the policy took effect. Subgroups analysis on part-time workers and workers with low earnings showed that these workers similarly experienced no change in their employment levels and small changes in their employment flows and volatility, echoing the impact estimates of workers overall. In my third study, I asked 1) What is the impact of the Seattle Minimum Wage Ordinance (MWO) on employment flows (hires, separations, turnover) in the low-wage market? 2) What is the impact of the Ordinance on hours volatility among continuing jobs? I defined the low-wage labor market as jobs with wages greater than 150% of the post-policy minimum wage – jobs that pay <$19 – because research has shown that cascading effects from minimum wage policies are less likely to occur beyond this wage threshold (Neumark, Schwizer, and Wascher, 2004). I used two quasi-experimental designs (synthetic control and interactive fixed-effect estimators) to estimate the impact of Seattle’s MWO on hires, separations, job turnover, and two measures of hours volatility among jobs which remained post-policy. I found that total separations and hires in the low-wage labor market declined following the minimum wage enactment, with the most precise reductions occurring during the period in which the minimum wage was raised to $13 per hour. Hires fell by a range of 12.2 to 19.3 percent. This decline in hires is met with a less precise, but persistent decline in separations, yielding no statistically significant change in job turnover among low-wage jobs in Seattle, relative to the comparison groups. Jobs that continued to exist in the post-policy period exhibited an increase in quarterly hours volatility ranging from 1.1 to 2.8 percent in the post-policy period. The majority of the treatment effects for the number of large (>25 percent) hours declines within a job were negative and statistically significant when the minimum wage increased up to $13 per hour. The timing of the decline in large hours drops coincided with a statistically significant decline in separations, indicating that the increases in hours volatility may be due to hours increases from continuing jobs.
Four Essays on Decentralized Markets in Management and Policy
(2020-02-04) Hayes, Adam Lee; Layton, David F
Governments often create markets to serve management or policy goals. The creation of these markets requires developing the institutional capacity to disseminate market information in order for the market to achieve those goals more effectively. This dissertation draws on prior work regarding network information diffusion to examine the ways in which information brokers affect market function and outcomes in management and policy contexts. The first two chapters focus on the U.S. municipal bond market, which serves as the management context, and uses data on all U.S. municipal bond issues for 2004-2016. The first chapter develops a measure of network connectivity for financial advisors in the market, and shows through linear models under nearest-neighbor matching that these highly connected advisors are associated with reduced interest costs for local governments. The second chapter uses a choice model under state dependence to estimate how local governments' past advisor choices implicitly constrain their future choice sets. The results suggest state dependence could increase interest costs for local governments, particularly for those that have contracted with only one or two advisors in the past. Alaska halibut and sablefish individual quota fisheries serve as the policy context for chapters 3 and 4 with confidential quota transaction data from AKFIN as the primary data source. In chapter 3, network model simulation shows that quota brokers help traders access the broader market by trading outside of their immediate social network compared to trading without a broker, though there still exists some relationship between traders' social networks and trading patterns even for brokered trades. Brokers are also shown to increase sales price across many quota sub-markets. Chapter 4 develops a novel choice-based method for estimating willingness-to-pay for quota among subgroups. This method is then applied this model within a regression discontinuity design to test how a recent policy change affected resource valuation among key subgroups of interest.
Consequences of Misassignment to Treatment: Examining Targeted Policy Interventions in Education
(2019-10-15) Kennedy, Alec Ito; Long, Mark C.
In education, students are often assigned to specialized programs or receive targeted interventions based on some standards and requirements. The criteria used to select the students for such targeted policies must weigh the trade-off between serving too many students and not serving enough. Under such selection criteria, I identify students who are selected into a treatment category that does not maximize their benefits as consequences of misassignment to treatment. As targeted policy interventions are aimed at providing supports for students with specific needs, it is important that we establish a firm understanding and provide guidance on how selection criteria should be designed and used to assure that all students are appropriately served. This dissertation explores several aspects of such selection criteria through the examination of three policies in education that seek to select students (or school districts) for targeted interventions and supports. The exploration provides learning and recommendations on how federal, state, and local governments can leverage data and analysis to limit the consequences of misassignment to treatment. The first chapter of this dissertation introduces a framework to facilitate in the understanding of the consequences to misassignment to treatment. The framework establishes a way of understanding how selection criteria for targeted policy interventions can fail and lead to losses of benefits for both targeted and non-targeted groups. The second chapter, Are We Correctly Measuring the Disproportionality of Minority Students in Special Education?, asks whether the current methods for measuring disproportionality in special education are accurately identifying school districts in need of a change of referral practices. Currently, the U.S. Department of Education's Office of Civil Rights (OCR) monitors whether school districts disproportionately refer minority students into special education services in an effort to prevent overreferral. However, these monitoring methods do not account for student-level factors (notably, socioeconomic status) which have been theorized to be positively associated with the presence of a learning disability. If minority students are more likely to experience low socioeconomic status than their White peers, then there may be good reason for minority students to be overrepresented in special education (i.e., the current measures of disproportionality could be susceptible to bias). In this chapter, I introduce an alternative measure for tracking disproportionality that removes this threat of bias. I then compare the performance of this alternative measure of disproportionality with the one currently in use and find that the currently used measure tends to overstate disproportionality. This finding suggests that current measures used by states to monitor disproportionality may incorrectly label some school districts as having an overrepresentation problem when, in fact, they do not. The third chapter, Successfully Transitioning English Language Learners into the Mainstream Classroom, investigates whether the state-established thresholds on language assessments used to determine whether English language learners (ELLs) are ready to transition from specialized services into the mainstream classroom are set in a way that maximizes the estimated benefits for ELLs in all instructional contexts. In this chapter, I find that the cutscores used to determine eligibility for ELL program exit in one state are generally set to promote the successful transition of ELLs into the mainstream classroom. However, I find that the cutscores do not work for all. Specifically, I find that Spanish-speaking students exiting bilingual programs struggle to succeed in English subjects after exiting ELL services. This finding suggests that the current thresholds are not working for all and should be modified in some way. This modification could involve the inclusion of teacher input into transitioning decisions or the close monitoring of recently exited ELL students in order to provide appropriate supports when needed. The fourth chapter, Making the Cut: An Optimization Approach for Setting Cutpoints in Targeted Policy Interventions in Education, provides an introduction to an optimization approach that can be used in Early Warning Indicators (EWI) or Early Warning Systems (EWS) to set optimal cutpoints on key leading indicators to determine eligibility for targeted supports. I apply the approach to two district examples of EWI/EWS to identify cutpoints that minimize misclassification rates. In the first example, I identify optimal cutpoints on kindergarten and first grade assessments to identify students who show evidence of having dyslexia and are in need of early literacy interventions. In the next example, I use the approach to produce thresholds that can be used to identify students at risk of becoming chronically absent in order to provide them with additional resources and support during the school year to help them overcome barriers to attendance. The approaches introduced and used in this chapter can offer guidance to districts or states seeking to establish EWI/EWS to identify students in need of targeted interventions or supports. The final chapter concludes this dissertation by presenting learning from each of the chapters on how selection criteria in targeted policy interventions can be better designed to limit consequences of misassignment to treatment.

Browse

Recent Submissions