The Impact of Medicaid Eligibility on SSI Participation: New Evidence from State-Funded Programs and Section 1115 Waivers

The Impact of Medicaid Eligibility on SSI Participation: New Evidence from State-Funded Programs and Section 1115 Waivers | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The Impact of Medicaid Eligibility on SSI Participation: New Evidence from State-Funded Programs and Section 1115 Waivers Yidonglin Liu, TianYi Zhu, Sixiao Liu, Chun-Chieh Hu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9667813/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This paper studies whether Medicaid expansions for low-income childless adults changed participation in Supplemental Security Income (SSI) before the Affordable Care Act. The expected direction of this relationship is theoretically ambiguous. Medicaid made available outside SSI can weaken the incentive to use SSI as a pathway to public insurance, but it can also increase SSI take-up by expanding contact with health providers and the welfare system, improving medical documentation, and revealing latent disability-related eligibility. Using data from the Current Population Survey and the Survey of Income and Program Participation from 2000 to 2013, we construct simulated Medicaid eligibility measures for state-funded programs and Section 1115 waivers and estimate the relationship using IV benchmarks and a special regressor estimator for binary outcome models with endogenous Medicaid coverage. The preferred special-regressor estimates indicate that Medicaid coverage induced by simulated eligibility increases SSI participation by 6.6–6.9 percentage points in the CPS and 5.3–6.4 percentage points in the SIPP. Effects are larger among unemployed individuals and are suggestively stronger among women in the CPS. The findings imply that Medicaid and SSI can operate as complements for economically vulnerable adults and show how expansions in one safety-net program may reshape participation in another. Health Economics & Outcomes Research Medicaid Supplemental Security Income simulated eligibility special regressor estimator welfare program interactions Figures Figure 1 Figure 2 Figure 3 Figure 4 1. Introduction The Supplemental Security Income (SSI) program occupies a central position in the U.S. safety net. It provides cash assistance to low-income individuals who are aged, blind, or disabled, while also serving as an important gateway to Medicaid in many states (Yelowitz 1998). This institutional linkage means that changes in Medicaid eligibility can reshape the incentives surrounding SSI participation. When Medicaid becomes available outside the SSI system, the value of applying for SSI may decline (Burns and Dague 2017). At the same time, broader Medicaid eligibility may increase contact with medical providers and public assistance agencies, potentially encouraging SSI take-up among individuals who were already close to program eligibility. Using CPS and SIPP data from 2000 to 2013, this paper finds that Medicaid coverage induced by simulated eligibility increases SSI participation by 5.3 to 6.9 percentage points. The result suggests that Medicaid and SSI can operate as complements rather than substitutes, implying that Medicaid expansion has spillovers across safety-net programs. This ambiguity motivates the policy question studied in this paper. Medicaid and SSI may operate as substitutes if expanded Medicaid coverage reduces the need to use SSI as a route to public health insurance. They may instead operate as complements if Medicaid eligibility increases awareness of public benefits (Bhargava and Manoli 2015; Finkelstein and Notowidigdo 2019), improves medical documentation, or helps individuals identify disability-related eligibility. The direction of the effect is therefore not determined mechanically by program rules. It depends on how low-income individuals respond to overlapping eligibility systems, application costs (Deshpande and Li 2019), health-care access, and labor-market constraints. The pre-ACA period provides a useful setting for studying this interaction. Before the ACA, low-income childless adults had limited access to Medicaid under federal rules, and coverage expansions depended heavily on state policy choices. Some states expanded coverage through state-funded programs, while others used Section 1115 waivers (Burns and Dague 2017; Dague et al. 2017; Bradley and Sabik 2019) to extend eligibility to groups that were otherwise excluded. These policy changes created substantial cross-state and over-time variation in Medicaid availability for childless adults, a group for whom SSI could otherwise be one of the most salient routes to public health insurance. Existing empirical evidence is mixed. Several studies emphasize a substitution channel (Burns and Dague 2017), finding that Medicaid expansions reduced SSI participation or SSI applications by lowering the need to obtain Medicaid through disability-based programs. Other evidence suggests that the relationship depends strongly on study design (Ne’eman and Maestas 2023), population (Staiger et al. 2024), and institutional setting (Baicker et al. 2014; Maestas et al. 2014; Schmidt et al. 2020). This paper adds to this literature by focusing on individual-level Medicaid coverage in a binary SSI participation model and using simulated eligibility to isolate policy-driven variation in Medicaid access. The empirical challenge is that Medicaid coverage is not randomly assigned. Individuals who obtain Medicaid may differ systematically from those who do not in health status, labor-market attachment, benefit knowledge, and underlying disability risk. These same unobserved factors may also affect SSI participation. To address this concern, the paper uses simulated Medicaid eligibility measures (Cutler and Gruber 1996; Currie and Gruber 1996) based on state-funded programs and Section 1115 waivers as instruments and applies the special regressor approach (Lewbel 2000; Dong and Lewbel 2015) for binary outcome models with endogenous regressors. This approach complements conventional IV-LPM and IV-Probit specifications by providing a semiparametric framework that relies less on linear probability assumptions. The main results indicate a positive relationship between Medicaid coverage and SSI participation. In the preferred special regressor estimates, Medicaid coverage induced by simulated eligibility increases SSI participation by 6.6 and 6.9 percentage points in the CPS under state-funded and Section 1115 waiver programs, respectively. The corresponding estimates in the SIPP are 5.3 and 6.4 percentage points. The effects are substantially larger among unemployed individuals, consistent with the view that Medicaid eligibility is more closely linked to SSI participation for adults with weaker labor-market attachment and lower earnings capacity (Autor and Duggan 2003; Maestas et al. 2013; Garthwaite et al. 2014). Gender heterogeneity is also present in the CPS, although it is less stable in the SIPP. The paper makes three contributions. First, it reframes the Medicaid-SSI relationship as an empirical question about substitution versus complementarity across welfare programs. Second, it distinguishes two pre-ACA policy channels-state-funded programs and Section 1115 waivers-that expanded Medicaid access for childless adults. Third, it applies the special regressor approach to a binary-outcome setting in which actual Medicaid coverage is endogenous, thereby complementing the existing DID and IV-LPM evidence. The results are therefore best interpreted as evidence on program interaction rather than as a claim that Medicaid mechanically raises SSI participation for all individuals. The findings indicate that, for low-income childless adults affected by Medicaid eligibility expansions, Medicaid access may increase the probability of SSI receipt through welfare-system contact (Bhargava and Manoli 2015), medical documentation (Messel et al. 2023), and latent disability-related eligibility (Deshpande and Li 2019). The paper proceeds as follows. Section 2 reviews the literature on Medicaid-SSI interactions and simulated eligibility. Section 3 describes the data, variable construction, and empirical strategy. Section 4 presents the baseline estimates, special regressor results, heterogeneity analyses, and robustness checks. Section 5 concludes by discussing the implications for the design of overlapping welfare programs. 2. Literature Review 2.1 Previous Research on SSI/SSDI and Medicaid A large literature studies interactions among SSI, SSDI, and Medicaid, with particular attention to whether public health insurance access changes incentives to apply for disability-related cash benefits. Early work (Yelowitz 1998) emphasized the value of Medicaid as a component of SSI participation and showed that the Medicaid link could affect the attractiveness of disability programs (Autor and Duggan 2003; Maestas et al. 2013). This perspective implies that expanding Medicaid outside SSI may reduce SSI participation by weakening the health-insurance motive to apply. Burns and Dague (2017) provide a key benchmark for this substitution view. Using pre-ACA Medicaid expansions for childless adults, they show that Medicaid coverage made available independently of SSI was associated with a decline in SSI participation among non-disabled childless adults aged 21–64. Staiger et al. (2024) similarly examine Medicaid expansion and SSI take-up using CPS ASEC data and highlight heterogeneity across racial and ethnic groups. These studies are important because they identify reduced-form policy effects of Medicaid expansions at the state level. Other studies find smaller, statistically insignificant, or more design-sensitive effects. Evidence from the Massachusetts health reform (Maestas et al. 2014), the Oregon Medicaid experiment (Baicker et al. 2014), and broader analyses of Medicaid expansions (Schmidt et al. 2020) suggests that estimates can vary depending on the population, disability category, timing, and empirical design. Ne’eman and Maestas (2023), in particular, emphasize that conclusions about Medicaid and disability-program participation are sensitive to study design. This sensitivity leaves room for complementary mechanisms, especially among individuals whose Medicaid coverage changes through eligibility rules and who may face high application costs (Deshpande and Li 2019) or limited knowledge of benefit programs (Bhargava and Manoli 2015). This paper builds on that debate by focusing on whether Medicaid eligibility can increase SSI participation among low-income childless adults. This positive channel is conceptually distinct from the standard substitution mechanism. Medicaid eligibility may increase medical contact (Messel et al. 2023), generate better documentation of health limitations, and expose individuals to administrative information about other public programs (Yelowitz 1998). If these forces dominate the reduction in SSI's health-insurance value, Medicaid and SSI may appear as complements for marginally affected individuals. 2.2 Simulated Eligibility (SE) Simulated eligibility has been widely used to separate policy variation in public insurance rules from individual-level selection into coverage. The central idea is to construct a measure of eligibility that reflects statutory program rules (Cutler and Gruber 1996; Currie and Gruber 1996) rather than individual choices or unobserved health demand. In Medicaid research, this approach is useful because actual Medicaid coverage may respond to income, employment, health status, and program knowledge, all of which may also be related to SSI participation. Bradley and Sabik (2019) use simulated eligibility to study Medicaid expansions and labor supply among childless adults. Their approach summarizes state-year variation in eligibility rules and uses that variation as an instrument for Medicaid coverage. Related work on public health insurance expansions (Cutler and Gruber 1996; Currie and Gruber 1996) similarly uses simulated eligibility to address endogenous population composition and to isolate policy-driven changes in program access. Following this literature, this paper constructs separate simulated eligibility measures for state-funded programs and Section 1115 waivers. This distinction is important because the two policy channels differed in financing, implementation, and eligibility thresholds. Estimating them separately allows the analysis to ask whether the Medicaid-SSI relationship is similar across different pre-ACA expansion mechanisms. 3. Data and Methodology 3.1 Data 3.1.1 State Medicaid programs In the United States, Medicaid eligibility before the ACA varied substantially by demographic group and state policy. Pregnant women, children, and some parents could qualify through federal and state eligibility categories, but low-income childless adults were generally less likely to be covered under standard federal rules. For this group, Medicaid access often depended on whether a state implemented a Section 1115 waiver (Burns and Dague 2017; Dague et al. 2017) or created a state-funded program (Bradley and Sabik 2019) outside the standard federal eligibility pathway. To capture this variation, the analysis follows Bradley and Sabik (2019) and uses Appendix Table 10 in Bradley and Sabik (2019) as the basis for the Medicaid policy rules. The policy information summarizes state-funded programs and Section 1115 waivers from 2000 to 2013, including implementation years and income thresholds expressed as percentages of the federal poverty level (FPL). Because eligibility thresholds often differed between working and non-working adults, the construction distinguishes between these groups when assigning simulated eligibility. Programs with limited benefits are excluded from the simulated eligibility calculation. Hawaii, Alaska, and the District of Columbia are also excluded because of program complexity, small cell sizes, or overlapping eligibility rules that make it difficult to assign a single policy channel. 3.1.2 March Current Population Survey (CPS) The first individual-level dataset is the March Current Population Survey (CPS), an annual household survey fielded by the U.S. Census Bureau. The CPS provides nationally representative information on demographic characteristics, labor-market status, income, and participation in public programs, including Medicaid and SSI. The CPS sample covers 2000–2013 and is restricted to childless adults aged 20–64, the population most directly affected by pre-ACA Medicaid expansions for childless adults. Individuals serving in the military and respondents reporting negative income are excluded to maintain consistency in eligibility measurement and sample composition. 3.1.3 Survey of Income and Program Participation (SIPP) The second dataset is the Survey of Income and Program Participation (SIPP), also administered by the U.S. Census Bureau. The SIPP provides detailed longitudinal information on income, employment, household characteristics, and program participation, making it useful for cross-validating patterns found in the CPS. The analysis uses SIPP core data from 2000 to 2013. Because the SIPP records program participation at a monthly frequency, the dataset contains richer within-year information than the CPS. This structure allows the analysis to examine whether the positive Medicaid-SSI relationship is robust in a longitudinal survey setting, while recognizing that CPS and SIPP differ in sampling design and measurement. 3.1.4 Dependent variable The dependent variable is SSI participation, defined as whether an individual receives Supplemental Security Income. In the CPS, SSI participation is measured using the SSI-YN item. The variable SSI equals one if the respondent reports receiving SSI and zero otherwise. In the SIPP, SSI participation is constructed using the variables RCUTYP03 and RCUTYP04, which identify federal and state SSI receipt. The variable SSI equals one if the respondent is covered by any SSI program and zero otherwise. 1 3.1.5 Independent variable The key endogenous explanatory variable is Medicaid coverage. In the CPS, Medicaid coverage is measured using the CAID indicator, which records whether the respondent is covered by Medicaid or a state-specific equivalent program. The variable Medicaid equals one if the respondent reports Medicaid coverage and zero otherwise. In the SIPP, Medicaid coverage is constructed from ECDMTH, which identifies whether the respondent is covered by Medicaid in a given month. The variable Medicaid equals one when the respondent reports Medicaid coverage in that month and zero otherwise. 3.1.6 Control variable The analysis includes individual and regional controls commonly used in the welfare and health economics literature (Autor and Duggan 2003; Dague et al. 2017; Bradley and Sabik 2019). Individual-level controls include gender, age and age squared, marital status, employment status, educational attainment, and race or ethnicity. Marital status and employment status are measured as binary indicators. Race and ethnicity are captured using indicators for White, Black, and Hispanic respondents. Regional controls include metropolitan status, the number of community hospitals in the state-year, and the natural logarithm of state GDP. 2 These variables account for local economic conditions, health-care supply, and regional differences that may be correlated with both Medicaid coverage and SSI participation. 3.2 Methodology 3.2.1 Simulated Eligibility (SE) A central empirical concern is the endogeneity of Medicaid coverage. Individuals who obtain Medicaid may differ from non-covered individuals in unobserved health, disability risk, program knowledge, and labor-market attachment. To address this issue, the paper uses simulated eligibility as an instrument (Cutler and Gruber 1996; Currie and Gruber 1996; Bradley and Sabik 2019). Simulated eligibility captures policy-induced variation in Medicaid access generated by state-funded programs and Section 1115 waivers, rather than relying solely on observed Medicaid coverage choices. To construct the simulated eligibility variables, the analysis uses childless adults aged 20–64 from the March CPS between 2000 and 2013. Family income is converted into a percentage of the FPL, and respondents are assigned eligibility according to the relevant state-year program rules in Appendix Table 10 in Bradley and Sabik (2019). Because some states operated both state-funded programs and Section 1115 waivers, the paper constructs two separate measures: se_state for state-funded programs and se_1115 for Section 1115 waivers. 3.2.2 Special Regressor Estimator In policy settings with a binary outcome and an endogenous binary explanatory variable, the instrumental-variable linear probability model is often used as a benchmark. However, the LPM may generate fitted probabilities outside the unit interval and can impose restrictive linearity assumptions. The special regressor estimator provides an alternative semiparametric approach for binary-choice models with endogenous regressors, following Lewbel (2000), Dong and Lewbel (2015), and related applications. The starting point is a binary-choice model in which the observed outcome equals one when the latent index is positive: $$\:D=I\left({\varvec{X}}^{\mathbf{{\prime\:}}}\beta\:+\epsilon\:\ge\:0\right)$$ where D represents the binary outcome variable, \(\:{\varvec{X}}^{\mathbf{{\prime\:}}}\) is the vector containing all observable regressors, \(\:\epsilon\:\) follows a zero-mean distribution, and \(\:I\left({\bullet\:}\right)\) represents the function that takes value one if the latent variable inside is positive, and zero otherwise. And to get the special regressor model, we need two more variables, an instrumental variable (Z) and a special regressor (V). And the special regressor should not be a part of the instrumental variables. In our study, the instrumental variable we choose is SE, for it satisfies the requirements of relevance and exogeneity. And we adopt the variable age as our special regressor (V), which meets the three criteria (Lewbel 2014). It is continuously distributed with wide support, because age ranges from 20 to 64. And it is also exogenous, for the sample’s age is predetermined with respect to the policy variation used to construct simulated eligibility. In the empirical sample, the special-regressor monotonicity condition requires \(\:E(D\mid\:\varvec{X},V)\) to be monotonic in \(\:V\) . This condition is plausible because SSI participation generally increases with age, and Fig. 1 shows a positive relationship between age and SSI participation in both datasets. Prior studies show that the probability of receiving welfare benefits such as SSI increases with age. And in both of our two datasets, a fitted kernel-weighted local polynomial regression of SSI participation on age provides evidence that there is a positive relationship between them (Fig. 1 ) (Autor and Duggan 2003; Maestas et al. 2013). These three properties make age a useful special regressor in this model. Following the estimation strategy summarized by Bontemps and Nauges (2016), the special regressor method transforms the binary outcome into a form that can be estimated using instrumental variables. The implementation proceeds in four steps. First, define the auxiliary variable T as follows: $$\:T=\frac{D-I(V\ge\:0)}{{f}_{V|Z}\left(V|Z\right)}$$ where \(\:{f}_{V|Z}\left(V\right|Z)\) presents the conditional probability density function of V given Z. Then, based on the assumptions, it could be shown that \(\:E\left(T\right|Z)=E({\varvec{X}}^{\mathbf{{\prime\:}}}\beta\:+\epsilon\:\left|Z\right)\) , where \(\:\epsilon\:\) is assumed independent of \(\:V|Z\) . This implies the following condition that \(\:E\left(Z{\varvec{X}}^{\mathbf{{\prime\:}}}\right)\beta\:=E\left(ZT\right)\) , which leads to the following expression for \(\:\beta\:\) . $$\:\beta\:={\left[E\left(\varvec{X}{Z}^{{\prime\:}}\right){E\left(\varvec{Z}{Z}^{{\prime\:}}\right)}^{-1}E\left(Z{\varvec{X}}^{\mathbf{{\prime\:}}}\right)\right]}^{-1}\hspace{0.25em}E\left(\varvec{X}{Z}^{{\prime\:}}\right){E\left(Z{Z}^{{\prime\:}}\right)}^{-1}E\left(ZT\right)$$ which is the definition of a linear two-stage least squares regression of \(\:T\) on \(\:\varvec{X}\) by using instrumental variable \(\:Z\) . To implement this estimator in practice, we proceed with the following steps: Step 1: The special regressor V is first centered. Then V is regressed on the observable covariates X and the instrument Z using ordinary least squares: $$\:{\widehat{U}}_{i}={V}_{i}-\left({\varvec{X}}^{\mathbf{{\prime\:}}}{\widehat{b}}_{X}+{Z}^{{\prime\:}}{\widehat{b}}_{Z}\right)$$ where \(\:{\widehat{b}}_{X}\) and \(\:{\widehat{b}}_{Z}\) are OLS estimated coefficients for variables in X and Z, and \(\:{\widehat{U}}_{i}\) are the residuals. Step 2: Using the residuals \(\:{\widehat{U}}_{i}\) , estimate their density via a kernel density estimator, which can be written as $$\:{\widehat{f}}_{h}\left(u\right)=\frac{1}{nh}\sum\:_{j=1}^{n}K\left(\frac{{\widehat{U}}_{j}-u}{h}\right)$$ where \(\:K\left({\bullet\:}\right)\) is a kernel function, and h is the bandwidth parameter. This yields a density estimate \(\:{\widehat{f}}_{i}=\widehat{f}\left({\widehat{U}}_{i}\right)\) . An alternative density estimator, the sorted-data estimator proposed by Lewbel and Schennach (2007) can also be used. Step 3: After obtaining the density estimator, \(\:{\widehat{T}}_{i}\) for each observation \(\:i\) could be calculated as: $$\:{\widehat{T}}_{i}=\frac{{D}_{i}-I\left({V}_{i}\ge\:0\right)}{{\widehat{f}}_{i}}$$ where the original binary outcome \(\:{D}_{i}\) is transformed into a continuous form suitable for later instrumental variable regression. Step 4: Finally, the transformed outcome is estimated using a two-stage least squares regression on the covariates, with the simulated eligibility measure serving as the instrument. The resulting estimate captures the effect of the endogenous Medicaid coverage variable under the special regressor assumptions. After identifying V, it can be added to the binary choice model, and \(\:{\varvec{X}}^{\mathbf{{\prime\:}}}\) can be divided into endogenous ( \(\:{\varvec{X}}^{\varvec{e}}\) ) and exogenous ( \(\:{\varvec{X}}^{0}\) ) covariates. The SR model can therefore be written as follows: $$\:D=I\left({\varvec{X}}^{e}{\beta\:}_{e}+{\varvec{X}}^{0}{\beta\:}_{0}+V+\epsilon\:\ge\:0\right)$$ In the empirical application, the model is estimated using a standard kernel density estimator. The robustness checks vary the kernel function and bandwidth. The estimating equation can be written as: $$\:{SSI}_{ist}=I\left({{M}_{ist}}^{e}{\beta\:}^{{\prime\:}}+{\varvec{X}}_{\varvec{i}\varvec{s}\varvec{t}}\gamma\:+V+{\epsilon\:}_{ist}\ge\:0\right)$$ Here, SSI denotes the binary indicator for SSI participation; Medicaid is the endogenous Medicaid coverage variable for individual i in state s and year t; X is the vector of exogenous controls; and age is the special regressor. Because coefficients in special regressor models do not have a direct probability interpretation, the analysis reports marginal effects. Following Dong and Lewbel (2015) and Bontemps and Nauges (2016), the marginal effects summarize how Medicaid coverage changes the probability of SSI participation under the estimated binary-choice model. 4. Results 4.1 Descriptives Table 1 defines the variables used in the empirical analysis. Because some states operated state-funded programs and Section 1115 waivers in the same period, the analysis constructs separate simulated eligibility measures for each policy channel. State GDP is transformed into logarithmic form to reduce the influence of extreme values and to capture differences in local economic conditions more flexibly. Table 2 presents descriptive statistics for the CPS and SIPP samples. In the CPS, the final sample includes 229,978 observations, with 183,728 observations in the Section 1115 waiver sample and 68,477 observations in the state-funded program sample. SSI participation is relatively low, at roughly 3 percent, while Medicaid coverage ranges from 8.75 to 10.16 percent across the two policy samples. These patterns are consistent with the fact that both programs target relatively disadvantaged low-income populations (Burns and Dague 2017; Bradley and Sabik 2019). The SIPP sample exhibits a similar structure. Simulated eligibility rates are approximately 21.8 percent for Section 1115 waivers and 21.5 percent for state-funded programs, while SSI participation remains much lower. The comparison between CPS and SIPP provides a useful robustness check because the two surveys differ in design, frequency, and measurement of program participation. 4.2 Baseline Regression 4.2.1 Results from the CPS Table 3 reports baseline LPM estimates for the CPS. These specifications should be interpreted as descriptive associations between Medicaid coverage and SSI participation before the instrumental-variable and special regressor estimates are introduced. Across specifications, the coefficient on Medicaid remains positive and statistically significant. In the preferred specification with controls and state and year fixed effects, Medicaid coverage is associated with a 27.0 percentage-point increase in SSI participation. Given the low baseline rate of SSI receipt in the sample, this is an economically large association. The magnitude reinforces the need to address endogeneity directly, since actual Medicaid coverage is likely correlated with unobserved health (Yelowitz 1998), disability (Autor and Duggan 2003; Maestas et al. 2013), and program participation propensities. 4.2.2 Results from the SIPP Table 4 reports the corresponding baseline estimates using the SIPP. The pattern is similar to the CPS: Medicaid coverage is positively associated with SSI participation across all four specifications. In the specification with controls and state and year fixed effects, the coefficient is 40.3 percentage points. These baseline results establish a strong positive association between Medicaid coverage and SSI participation. Because they do not by themselves resolve the endogeneity of Medicaid coverage, the next section turns to instrumental-variable and special regressor estimates. 4.3 Estimation Results This section compares the special regressor estimates with conventional IV-Probit and IV-LPM benchmarks. The comparison is useful because the outcome is binary and the treatment variable, Medicaid coverage, is potentially endogenous. Table 5 reports estimates from IV-Probit, IV-LPM, and the special regressor model (Lewbel 2000; Dong and Lewbel 2015; Bontemps and Nauges 2016). Because IV-Probit and special regressor coefficients are not directly interpretable as probability effects, the discussion focuses on the average marginal effects. The IV-Probit estimates provide a nonlinear benchmark, while the IV-LPM estimates provide a conventional linear probability benchmark. The special regressor estimates are the preferred semiparametric estimates because they address the endogeneity of Medicaid coverage in a binary-outcome framework without relying on the linear probability specification. In the CPS, the IV-Probit marginal effects are 7.0 percentage points under the state-funded instrument and 7.4 percentage points under the Section 1115 waiver instrument. These estimates are very close to the special regressor marginal effects, which are 6.6 and 6.9 percentage points, respectively. By contrast, the IV-LPM estimates are substantially larger, at 24.3 and 25.3 percentage points. This pattern suggests that the linear probability framework may overstate the magnitude of the relationship between Medicaid coverage and SSI participation in this binary-outcome setting. A similar pattern appears in the SIPP. The IV-Probit marginal effects are 7.6 percentage points under the state-funded instrument and 9.1 percentage points under the Section 1115 waiver instrument. The corresponding special regressor marginal effects are 5.3 and 6.4 percentage points. As in the CPS, the IV-LPM estimates are much larger, at 38.4 and 35.4 percentage points. Taken together, the results indicate that Medicaid coverage induced by simulated eligibility is positively associated with SSI participation across both datasets and both policy instruments. The magnitude of the effect is more moderate under the nonlinear and semiparametric specifications than under the linear probability model. Overall, the estimates in Table 5 support the interpretation that Medicaid and SSI may operate as complements for low-income childless adults in this setting (Yelowitz 1998). Rather than simply reducing the value of SSI as a pathway to Medicaid, Medicaid coverage may increase contact with the public assistance system (Bhargava and Manoli 2015), improve access to medical documentation (Messel et al. 2023), and reveal latent eligibility for disability-related benefits (Deshpande and Li 2019). This interpretation is consistent with the positive and stable marginal effects from the IV-Probit and special regressor estimates, while the larger IV-LPM estimates should be interpreted more cautiously as linear probability benchmarks. 4.4 Heterogeneity Analysis: Gender and Employment Status Building on prior evidence that Medicaid effects vary by demographic group and study design (Staiger et al. 2024; Ne’eman and Maestas 2023), the analysis next examines heterogeneity by gender and employment status. The subgroup estimates use the IV-LPM specification from Table 5 and are reported separately for the CPS and SIPP. Table 7 shows that Medicaid coverage has a positive and statistically significant effect for both men and women. In the CPS, the estimated effects are larger for women than for men under both state-funded programs and Section 1115 waivers. In the SIPP, women experience a larger effect under state-funded programs, while the gender difference is less pronounced under Section 1115 waivers. This pattern suggests that gender heterogeneity is present but should be interpreted cautiously because it is not fully stable across datasets. Table 8 shows a more robust pattern by employment status. In both the CPS and SIPP, the estimated effect is substantially larger among unemployed individuals than among employed individuals. For example, in the CPS state-funded specification, Medicaid coverage increases SSI participation by 30.7 percentage points among unemployed individuals, compared with 9.9 percentage points among employed individuals. The same qualitative pattern appears under Section 1115 waivers and in the SIPP. The employment heterogeneity is consistent with the view that Medicaid eligibility is especially relevant for individuals with weaker labor-market attachment and lower earnings capacity (Garthwaite et al. 2014; Dague et al. 2017). For this group, Medicaid coverage may increase interaction with the welfare system and may also coincide with characteristics that make SSI participation more likely. The gender results are more suggestive, whereas the employment-status results are consistent across both datasets and policy channels. 4.4 Robustness Checks 4.4.1 Placebo Tests The analysis conducts two placebo exercises to examine whether the main results are likely to be driven by broad state-level trends or random assignment of policy exposure. The first placebo test restricts the sample to high-income childless adults, who are unlikely to be affected by Medicaid eligibility expansions targeted at low-income populations. Table 9 shows that the estimated marginal effects for the high-income group are all statistically insignificant. In particular, CPS estimates are close to zero, while SIPP estimates are imprecise and statistically insignificant. This pattern supports the interpretation that the main estimates are concentrated among the intended low-income population rather than reflecting general changes in SSI participation unrelated to Medicaid eligibility. The second placebo exercise replaces the true simulated eligibility instruments with randomly generated false eligibility measures over 1,000 simulations. Figures 2 and 3 show that the resulting placebo t-statistics are concentrated around zero and rarely exceed the conventional 5 percent critical value, while the actual estimates lie far in the right tail of the placebo distribution. This pattern indicates that the estimated relationship is driven by actual Medicaid eligibility variation rather than random assignment. 4.4.3 Sensitivity Tests The analysis also examines the sensitivity of the special regressor estimates (Lewbel and Schennach 2007; Bontemps and Nauges 2016) to the choice of kernel function. Table 10 reports estimates that replace the default Epanechnikov kernel with rectangle and triangle kernels under the default bandwidth. Across both datasets and both simulated eligibility instruments, the marginal effects remain positive and statistically significant, and the magnitudes are close to the preferred estimates. As an additional sensitivity check, Table 11 varies the bandwidth parameter of the default kernel from 0.1 to 1.0. Figure 4 plots the resulting marginal effects. The estimates remain positive across all bandwidths. In the CPS, the marginal effects are highly stable. In the SIPP, the state-funded estimates decline modestly as the bandwidth increases, while the Section 1115 estimates remain nearly flat. Overall, the bandwidth exercise indicates that the main conclusion does not depend on a particular smoothing parameter. 5. Conclusion This paper examines whether Medicaid eligibility expansions for low-income childless adults affected SSI participation before the Affordable Care Act. Using CPS and SIPP data from 2000 to 2013, simulated eligibility instruments based on state-funded programs and Section 1115 waivers, and a special regressor estimator for binary outcome models with endogenous regressors, the analysis finds a positive relationship between Medicaid coverage and SSI participation. The preferred estimates suggest that Medicaid coverage induced by simulated eligibility increases SSI participation by approximately 5 to 7 percentage points across datasets and policy measures. This finding differs from a simple substitution view (Burns and Dague 2017; Baicker et al. 2014) in which expanding Medicaid outside SSI necessarily reduces SSI participation by lowering the value of SSI as a route to health insurance. Instead, the results suggest that Medicaid and SSI may operate as complements for some low-income childless adults (Yelowitz 1998). The heterogeneity results provide additional support for this interpretation. The effect is consistently stronger among unemployed individuals, a group with weaker labor-market attachment and lower earnings capacity (Garthwaite et al. 2014; Dague et al. 2017). The gender pattern is more mixed: women experience larger effects in the CPS and under the SIPP state-funded specification, but the difference is less stable under Section 1115 waivers in the SIPP (Staiger et al. 2024). Methodologically, the paper complements conventional LPM, Probit, and IV specifications by applying the special regressor approach (Lewbel 2000; Dong and Lewbel 2015; Bontemps and Nauges 2016) to a setting with both a binary outcome and endogenous Medicaid coverage. The special regressor estimates are smaller than the baseline and IV-LPM estimates but remain positive across datasets, instruments, kernels, and bandwidth choices. Substantively, the findings highlight the importance of coordination across overlapping welfare programs. Expanding Medicaid eligibility may not only change health-insurance coverage; it may also alter participation in cash assistance programs by increasing contact with medical providers (Messel et al. 2023), improving documentation, and exposing individuals to information about related benefits (Bhargava and Manoli 2015; Finkelstein and Notowidigdo 2019; Deshpande and Li 2019). These channels imply that the fiscal and behavioral consequences of Medicaid expansions can extend beyond the Medicaid program itself. The analysis has several limitations. The CPS and SIPP differ in survey design and program-participation measurement, and the data do not directly observe the application process, medical documentation, or caseworker interactions. Future research using administrative application records (Deshpande and Li 2019) or health-care utilization data (Messel et al. 2023) could more directly distinguish the awareness, diagnosis, and labor-supply channels underlying the Medicaid-SSI relationship. Declarations Conflict of interest disclosure: No potential conflict of interest was disclosed by the author(s). Funding statement: This research did not receive any specific grant from a funding agency. Acknowledgement: None. Data availability statement: The data in this study can be requested from the corresponding author. References Autor, D.H., Duggan, M.G., 2003. The rise in the disability rolls and the decline in unemployment. The Quarterly Journal of Economics 118, 157–206. doi:10.1162/00335530360535171. Baicker, K., Finkelstein, A., Song, J., Taubman, S., 2014. The impact of Medicaid on labor market activity and program participation: Evidence from the Oregon health insurance experiment. American Economic Review 104, 322–328. doi:10.1257/aer.104.5.322. Bhargava, S., Manoli, D., 2015. Psychological frictions and the incomplete take-up of social benefits: Evidence from an IRS field experiment. American Economic Review 105, 3489–3529. doi:10.1257/aer.20121493. Bontemps, C., Nauges, C., 2016. The impact of perceptions in averting-decision models: An application of the special regressor method to drinking water choices. American Journal of Agricultural Economics 98, 297–313. doi:10.1093/ajae/aav046. Bradley, C.J., Sabik, L.M., 2019. Medicaid expansions and labor supply among low-income childless adults: Evidence from 2000 to 2013. International Journal of Health Economics and Management 19, 235–272. doi:10.1007/s10754-018-9248-x. Burns, M., Dague, L., 2017. The effect of expanding Medicaid eligibility on supplemental security income program participation. Journal of Public Economics 149, 20–34. doi:10.1016/j.jpubeco.2017.03.004. Currie, J., Gruber, J., 1996. Health insurance eligibility, utilization of medical care, and child health. The Quarterly Journal of Economics 111, 431–466. doi:10.2307/2946684. Cutler, D.M., Gruber, J., 1996. Does public insurance crowd out private insurance? The Quarterly Journal of Economics 111, 391–430. doi:10.2307/2946683. Dague, L., DeLeire, T., Leininger, L., 2017. The effect of public insurance coverage for childless adults on labor supply. American Economic Journal: Economic Policy 9, 124–154. doi:10.1257/pol.20150059. Deshpande, M., Li, Y., 2019. Who is screened out? application costs and the targeting of disability programs. American Economic Journal: Economic Policy 11, 213–248. doi:10.1257/pol.20180076. Dong, Y., Lewbel, A., 2015. A simple estimator for binary choice models with endogenous regressors. Econometric Reviews 34, 82–105. doi:10.1080/07474938.2014.944470. Finkelstein, A., Notowidigdo, M.J., 2019. Take-up and targeting: Experimental evidence from SNAP. The Quarterly Journal of Economics 134, 1505–1556. doi:10.1093/qje/qjz013. Garthwaite, C., Gross, T., Notowidigdo, M.J., 2014. Public health insurance, labor supply, and employment lock. The Quarterly Journal of Economics 129, 653–696. doi:10.1093/qje/qju005. Lewbel, A., 2000. Semiparametric qualitative response model estimation with unknown heteroscedasticity or instrumental variables. Journal of Econometrics 97, 145–177. doi:10.1016/S0304-4076(00)00015-4. Lewbel, A., 2014. An overview of the special regressor method, in: Ullah, A., Racine, J.S., Su, L. (Eds.), The Oxford Handbook of Applied Nonparametric and Semiparametric Econometrics and Statistics. Oxford University Press, Oxford, pp. 38–62. Lewbel, A., Schennach, S.M., 2007. A simple ordered data estimator for inverse density weighted expectations. Journal of Econometrics 136, 189–211. doi:10.1016/j.jeconom.2005.08.005. Maestas, N., Mullen, K.J., Strand, A., 2013. Does disability insurance receipt discourage work? using examiner assignment to estimate causal effects of SSDI receipt. American Economic Review 103, 1797–1829. doi:10.1257/aer.103.5.1797. Maestas, N., Mullen, K.J., Strand, A., 2014. Disability insurance and health insurance reform: Evidence from Massachusetts. American Economic Review 104, 329–335. doi:10.1257/aer.104.5.329. Messel, M., Swensen, I., Urban, C., 2023. The effects of expanding access to mental health services on SS(D)I applications and awards. Labour Economics 81, 102339. doi:10.1016/j.labeco.2023.102339. Ne'eman, A., Maestas, N., 2023. How Does Medicaid Expansion Impact Income Support Program Participation and Employment for Different Types of People with Disabilities? NBER Working Paper 31816. National Bureau of Economic Research. doi:10.3386/w31816. Sanderson, E., Windmeijer, F., 2016. A weak instrument F-test in linear IV models with multiple endogenous variables. Journal of Econometrics 190, 212–221. doi:10.1016/j.jeconom.2015.06.004. Schmidt, L., Shore-Sheppard, L.D., Watson, T., 2020. The impact of the ACA Medicaid expansion on disability program applications. American Journal of Health Economics 6, 444–476. doi:10.1086/710525. Staiger, B., Helfer, M.S., Van Parys, J., 2024. The effect of Medicaid expansion on the take-up of disability benefits by race and ethnicity. Health Economics 33, 526–540. doi:10.1002/hec.4783. Yelowitz, A.S., 1998. Why did the SSI-disabled program grow so much? disentangling the effect of Medicaid. Journal of Health Economics 17, 321–349. doi:10.1016/S0167-6296(97)00024-6. Footnotes In the SIPP, SSI receipt can be separately constructed for the policy-channel specifications. Therefore, the SIPP regressions use the SSI measure corresponding to the Section 1115 waiver and state-funded program specifications, respectively. Table 2 reports these SIPP SSI measures separately, while the CPS reports a single SSI participation measure because the CPS does not provide the same separate SSI classification. The number of community hospitals is obtained from the 1999–2024 AHA Annual Survey. State GDP is from the U.S. Bureau of Economic Analysis SASUMMARY state annual summary statistics. Tables Table 1 Variable Definitions Variable Definition SSI Whether the respondent receives Supplemental Security Income Medicaid Whether the respondent is covered by Medicaid SE_state Whether the respondent is eligible to receive Medicaid coverage through state-funded programs SE_1115 Whether the respondent is eligible to receive Medicaid coverage through Section 1115 waiver programs Age The age of the respondent Female Whether the respondent’s gender of birth is female HighEdu Whether the respondent has attained a degree higher than high school Work Whether the respondent is currently employed Married Whether the respondent is currently married MSA Whether the respondent currently lives in a metropolitan area lnGDP Natural logarithm of state GDP Hospital Number of community hospitals in the state-year White Whether the respondent belongs to the White racial category Black Whether the respondent belongs to the Black racial category Hispanic Whether the respondent belongs to the Hispanic racial category Notes : This table reports the definitions of the main variables used in the empirical analysis. For binary variables, a value of 1 indicates that the condition described in the definition is satisfied, and a value of 0 indicates otherwise. SSI refers to Supplemental Security Income participation, and Medicaid refers to Medicaid coverage. Table 2 Descriptive Statistics Variable SIPP CPS SE (1115) SE (state) SE (1115) SE (state) (21.5%) (21.8%) (19.44%) (19.02%) SSI* 4.20% 3.80% 2.99% 2.94% SSI (1115)* 4.20% 3.80% – – SSI (State)* 0.79% 0.56% – – Medicaid* 10% 8.60% 10.17% 8.75% Age 45 45 44 44 Female* 51% 51% 49% 49% Married* 47% 49% 48% 50% MSA* 80% 81% 80% 81% HighEdu* 66% 66% 58% 59% Work* 75% 77% 77% 80% Hospital 162 116 160 115 lnGDP 13 13 12 12 Race Black* 6.80% 4.80% 9.86% 7.47% White* 82% 89% 79.83% 84.25% Hispanic* 12% 5.40% 17.42% 10.82% Number of Obs. 630724 385293 183728 68477 Notes : Variables marked with * are binary variables. The table reports means for continuous variables and the proportion equal to 1 for binary variables. Columns labeled SE (1115) and SE (state) refer to the samples used for the Section 1115 waiver and state-funded program analyses, respectively. SSI and Medicaid are measured as participation or coverage indicators. The number of observations varies across columns because the state-funded and Section 1115 waiver samples are defined by different policy variation. Table 3 CPS Baseline Regression Results Variables (1) (2) (3) (4) Medicaid 0.295*** 0.268*** 0.297*** 0.270*** (0.003) (0.003) (0.003) (0.003) Age 0.001*** 0.001*** (0.000) (0.000) Female -0.003*** -0.003*** (0.001) (0.001) HighEdu -0.008*** -0.007*** (0.001) (0.001) Work -0.052*** -0.052*** (0.001) (0.001) Married -0.017*** -0.017*** (0.001) (0.001) MSA 0.003*** 0.002* (0.001) (0.001) Hospital 0.000*** 0.000*** (0.000) (0.000) lnGDP -0.001** 0.017 (0.001) (0.012) White 0.005*** 0.004*** (0.001) (0.001) Black 0.009*** 0.009*** (0.002) (0.002) Hispanic -0.008*** -0.005*** (0.001) (0.001) Constant 0.001*** 0.037*** 0.001*** -0.249* (0.000) (0.007) (0.000) (0.147) State FE NO NO YES YES Year FE NO NO YES YES Observations 230,314 229,978 230,314 229,978 R-squared 0.264 0.285 0.266 0.287 Notes : The dependent variable is SSI participation. Column (1) includes neither control variables nor fixed effects. Column (2) includes individual- and state-level controls. Column (3) includes state and year fixed effects. Column (4) includes both the full set of controls and state and year fixed effects and is the preferred baseline specification. Robust standard errors are in parentheses. \(\:{}^{\text{*}}p<0.10\) , \(\:{}^{\text{*}\text{*}}p<0.05\) , \(\:{}^{\text{*}\text{*}\text{*}}p<0.01\) . Table 4 SIPP Baseline Regression Results Variables (1) (2) (3) (4) Medicaid 0.403*** 0.402*** 0.405*** 0.403*** (0.001) (0.001) (0.001) (0.001) Age 0.001*** 0.001*** (0.000) (0.000) Female -0.008*** -0.008*** (0.000) (0.000) HighEdu -0.017*** -0.017*** (0.000) (0.000) Work 0.005*** 0.004*** (0.000) (0.000) Married -0.006*** -0.006*** (0.000) (0.000) MSA -0.046*** -0.046*** (0.000) (0.000) Hospital 0.000*** 0.000*** (0.000) (0.000) lnGDP -0.004*** 0.001 (0.000) (0.005) White 0.005*** 0.006*** (0.001) (0.001) Black 0.002*** 0.003*** (0.000) (0.000) Hispanic -0.012*** -0.012*** (0.001) (0.001) Constant 0.002*** 0.064*** 0.002*** -0.011 (0.000) (0.003) (0.000) (0.062) State FE NO NO YES YES Year FE NO NO YES YES Observations 1,756,294 1,581,626 1,756,294 1,581,626 R-squared 0.363 0.404 0.364 0.405 Notes : The dependent variable is SSI participation. Column (1) includes neither control variables nor fixed effects. Column (2) includes individual- and state-level controls. Column (3) includes state and year fixed effects. Column (4) includes both the full set of controls and state and year fixed effects and is the preferred baseline specification. Robust standard errors are in parentheses. \(\:{}^{\text{*}}p<0.10\) , \(\:{}^{\text{*}\text{*}}p<0.05\) , \(\:{}^{\text{*}\text{*}\text{*}}p<0.01\) . Table 5 Estimated Marginal Effects Estimation method Coefficient Average marginal effect Coefficient SE Panel A: CPS IV - Probit (state) 1.480*** 0.070 0.319 IV - Probit (1115) 1.897*** 0.074 0.177 IV - LPM (state) 0.243*** 0.243 0.017 IV - LPM (1115) 0.253*** 0.253 0.010 Kdens (state) 44.54*** 0.066 4.023 Kdens (1115) 41.19*** 0.069 2.454 Panel B: SIPP IV - Probit (state) 2.271*** 0.076 0.110 IV - Probit (1115) 2.441*** 0.091 0.071 IV - LPM (state) 0.384*** 0.384 0.007 IV - LPM (1115) 0.354*** 0.354 0.005 Kdens (state) 112.6*** 0.053 17.50 Kdens (1115) 127.9*** 0.064 28.25 Notes : IV-Probit coefficients are reported on the latent-index scale and are not directly interpretable as probability effects. For IV-LPM, the coefficient is directly interpretable as a probability effect. For the special regressor estimates, Kdens denotes the estimator using the default Epanechnikov kernel density method; its coefficients are not directly interpretable, so average marginal effects are reported separately. The reported standard errors refer to the estimated coefficients, and significance stars are based on coefficient estimates and their corresponding standard errors. \(\:{}^{\text{*}}p<0.10\) , \(\:{}^{\text{*}\text{*}}p<0.05\) , \(\:{}^{\text{*}\text{*}\text{*}}p<0.01\) . Table 6 First-Stage Regression Results of IV-LPM Variables CPS SIPP state 1115 state 1115 SE_state 0.126*** 0.137*** (0.004) (0.002) SE_1115 0.130*** 0.147*** (0.002) (0.001) Observations 68,477 183,728 385,293 630,724 State FE YES YES YES YES Year FE YES YES YES YES SW F 999.57 2732.93 7424.04 14557.68 SW Chi-sq 1000.02 2733.57 7424.65 14558.62 Notes : The dependent variable is Medicaid coverage. SE_state and SE_1115 are the excluded instruments used in the IV-LPM specifications. The Sanderson-Windmeijer first-stage chi-squared statistic (SW Chi-sq) and F statistic (SW F) report underidentification and weak-identification diagnostics for the endogenous Medicaid variable (Sanderson and Windmeijer 2016). All specifications include state and year fixed effects. Robust standard errors are in parentheses. \(\:{}^{\text{*}}p<0.10\) , \(\:{}^{\text{*}\text{*}}p<0.05\) , \(\:{}^{\text{*}\text{*}\text{*}}p<0.01\) . Table 7 Subgroup Analysis: Gender Male Female Panel A: CPS N 34,792 33,685 State and Year FEs with CV (state) 0.216*** 0.258*** (0.029) (0.021) N 93,116 90,612 State and Year FEs with CV (1115) 0.237*** 0.260*** (0.018) (0.013) Panel B: SIPP N 189,873 195,420 State and Year FEs with CV (state) 0.331*** 0.407*** (0.014) (0.008) N 308,202 322,522 State and Year FEs with CV (1115) 0.365*** 0.351*** (0.009) (0.006) Notes : Each coefficient is estimated from a separate IV-LPM regression using the specification in Table 5 . The sample is split by gender as indicated in the column header. The state-funded and Section 1115 waiver estimates are reported separately for the CPS and SIPP samples. Robust standard errors are in parentheses. \(\:{}^{\text{*}}p<0.10\) , \(\:{}^{\text{*}\text{*}}p<0.05\) , \(\:{}^{\text{*}\text{*}\text{*}}p<0.01\) . Table 8 Subgroup Analysis: Employment Status Unemployed Employed Panel A: CPS N 13,682 54,795 State and Year FEs with CV (state) 0.307*** 0.099*** (0.036) (0.015) N 42,352 141,878 State and Year FEs with CV (1115) 0.343*** 0.086*** (0.019) (0.009) Panel B: SIPP N 88,451 296,842 State and Year FEs with CV (state) 0.434*** 0.199*** (0.012) (0.009) N 160,185 470,539 State and Year FEs with CV (1115) 0.486*** 0.068*** (0.008) (0.005) Notes : Each coefficient is estimated from a separate IV-LPM regression using the specification in Table 5 . The sample is split by employment status as indicated in the column header. The state-funded and Section 1115 waiver estimates are reported separately for the CPS and SIPP samples. Robust standard errors are in parentheses. \(\:{}^{\text{*}}p<0.10\) , \(\:{}^{\text{*}\text{*}}p<0.05\) , \(\:{}^{\text{*}\text{*}\text{*}}p<0.01\) . Table 9 Placebo Test: High-Income Group Coefficient Marginal effect Standard Error Panel A: High-income group in CPS State-funded -26.43 -0.001 51.672 Section 1115 waiver -1.31 -0.000 50.931 Panel B: High-income group in SIPP State-funded 211.5 0.011 132.1 Section 1115 waiver -2306.9 0.099 1891.2 Notes : The high-income group is defined as individuals with total family income at or above the 60th percentile within each sample year. The table reports placebo estimates from the special regressor model. Marginal effects are average marginal effects on SSI participation. Reported standard errors refer to the estimated coefficients. \(\:{}^{\text{*}}p<0.10\) , \(\:{}^{\text{*}\text{*}}p<0.05\) , \(\:{}^{\text{*}\text{*}\text{*}}p<0.01\) . Table 10 Special Regressor Model Results with Different Kernel Functions Data Type of kernel function Average marginal effect Coefficient SE Panel A: CPS State-funded rectangle 0.066*** 2.426 triangle 0.064*** 2.394 Section 1115 waiver rectangle 0.062*** 1.447 triangle 0.063*** 1.435 Panel B: SIPP State-funded rectangle 0.072*** 21.71 triangle 0.064*** 21.74 Section 1115 waiver rectangle 0.066*** 28.44 triangle 0.065*** 28.71 Notes : Rectangle and triangle are alternative kernel functions used in the special regressor model. Each regression uses the Kdens specification in Table 5 with a different kernel function. Marginal effects are average marginal effects on SSI participation. Reported standard errors refer to the estimated coefficients. \(\:{}^{\text{*}}p<0.10\) , \(\:{}^{\text{*}\text{*}}p<0.05\) , \(\:{}^{\text{*}\text{*}\text{*}}p<0.01\) . Table 11 Special Regressor Model Results with Different Bandwidths Data Bandwidth State-funded ME State-funded SE Section 1115 ME Section 1115 SE CPS 0.10 0.064*** 2.287 0.061*** 1.407 0.20 0.065*** 2.400 0.061*** 1.418 0.30 0.064*** 2.409 0.062*** 1.448 0.40 0.064*** 2.395 0.062*** 1.439 0.50 0.064*** 2.401 0.062*** 1.437 0.60 0.064*** 2.400 0.062*** 1.440 0.70 0.064*** 2.399 0.062*** 1.437 0.80 0.064*** 2.401 0.063*** 1.437 0.90 0.064*** 2.399 0.063*** 1.439 1.00 0.064*** 2.397 0.062*** 1.437 SIPP 0.10 0.079*** 26.89 0.063*** 28.95 0.20 0.075*** 23.85 0.064*** 30.44 0.30 0.072*** 21.29 0.065*** 28.38 0.40 0.068*** 21.09 0.066*** 29.23 0.50 0.066*** 21.41 0.066*** 29.26 0.60 0.062*** 21.92 0.065*** 28.73 0.70 0.063*** 21.89 0.065*** 28.60 0.80 0.064*** 21.69 0.065*** 28.81 0.90 0.064*** 21.64 0.065*** 29.02 1.00 0.063*** 21.57 0.065*** 28.88 Notes : The special regressor model uses the default Epanechnikov kernel. Each specification uses a different bandwidth from 0.1 to 1.0. The bandwidth controls the degree of smoothing in the kernel density estimation: smaller bandwidths place more weight on nearby observations, while larger bandwidths produce smoother estimates. Marginal effects are average marginal effects on SSI participation. Reported standard errors refer to the estimated coefficients. \(\:{}^{\text{*}}p<0.10\) , \(\:{}^{\text{*}\text{*}}p<0.05\) , \(\:{}^{\text{*}\text{*}\text{*}}p<0.01\) . Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9667813","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":637561810,"identity":"f4e75b3f-9fba-42d2-9d17-41e44470d640","order_by":0,"name":"Yidonglin Liu","email":"","orcid":"","institution":"The University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Yidonglin","middleName":"","lastName":"Liu","suffix":""},{"id":637561811,"identity":"ac572528-e0dc-406b-9b22-1eedc109384f","order_by":1,"name":"TianYi Zhu","email":"","orcid":"","institution":"University of South California","correspondingAuthor":false,"prefix":"","firstName":"TianYi","middleName":"","lastName":"Zhu","suffix":""},{"id":637561812,"identity":"604cbced-fbec-4d5c-9812-28e3bb80d1ab","order_by":2,"name":"Sixiao Liu","email":"","orcid":"","institution":"The Chinese University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Sixiao","middleName":"","lastName":"Liu","suffix":""},{"id":637561813,"identity":"bcd5a97e-1d76-411f-b9b8-bf3f0d6404e1","order_by":3,"name":"Chun-Chieh Hu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/0lEQVRIiWNgGAWjYDACCTB5gAfIYGNgqACxYVIHsOtA03KGBC0MYC2MbURo4Z/dY/aAoeaODL90A9tj3nmH5fmO9x6T+LiHQY7vRgJ2S+6cMTdgOPaMR3LOAXZj3m2HDWeeOZcmOeMZg7EkDi0GEjlm0n8bDvMY3EhgkwZqSTC4kWNszHOAIXEDHi0SjEAt9mAtc4Ba7r8xNv5zgKGeoBYDCZCWBpAtPIaPgV4HMnD45UZamQTDscM8EjcS2yTnHEsH+iXH8GHPAQkg4wH2EJuRvE2CoeawPZBxTOJNjTUwxM4YHPhxwAbIwG4LEmBsQLGekPJRMApGwSgYBXgAAAvBX2boLG/wAAAAAElFTkSuQmCC","orcid":"","institution":"University of Colorado Denver","correspondingAuthor":true,"prefix":"","firstName":"Chun-Chieh","middleName":"","lastName":"Hu","suffix":""}],"badges":[],"createdAt":"2026-05-10 05:56:06","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9667813/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9667813/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109197419,"identity":"bffd5225-27d0-406d-b25d-16d01647c538","added_by":"auto","created_at":"2026-05-13 13:16:37","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":416222,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eKernel-weighted local polynomial regression of SSI participation on age\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNotes: The figure plots kernel-weighted local polynomial estimates of SSI participation against age, the special regressor used in the model. The shaded areas show 95 percent confidence intervals. The bandwidth is set to 4.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9667813/v1/81f0148734f4b810eaf4d2d1.jpeg"},{"id":109197422,"identity":"5d658b7b-8651-41c7-9bc5-eed0d0c4a95e","added_by":"auto","created_at":"2026-05-13 13:16:38","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":826671,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDistribution of simulated placebo t-values in CPS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNotes: The figure shows the distribution of 1,000 placebo simulations of the estimated t-values for predicted Medicaid eligibility in the CPS. The vertical solid line marks t = 1.96, corresponding to the 5% significance level. In each simulation, the true simulated eligibility (either the state-funded instrument, se_state, or the Section 1115 waiver instrument, se_1115) is replaced by a randomly generated false simulated eligibility. The resulting placebo t-value distributions for se_state and se_1115 are nearly the same. For ease of illustration, the figure reports only the distribution based on the state-funded instrument. For comparison, the actual t-value of Medicaid is 14.759 using the state-funded instrument and 25.881 using the Section 1115 waiver instrument.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9667813/v1/6a899f93a636ff5962cfbb22.png"},{"id":109205456,"identity":"c91e7911-b7d4-43e6-b7b1-7e024630cdb4","added_by":"auto","created_at":"2026-05-13 15:04:47","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":837313,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDistribution of simulated placebo t-values in SIPP\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNotes: The figure shows the distribution of 1,000 placebo simulations of the estimated t-values for predicted Medicaid eligibility in the SIPP. The vertical solid line marks t = 1.96, corresponding to the 5% significance level. In each simulation, the true simulated eligibility (either the state-funded instrument, se_state, or the Section 1115 waiver instrument, se_1115) is replaced by a randomly generated false simulated eligibility. The resulting placebo t-value distributions for se_state and se_1115 are nearly the same. For ease of illustration, the figure reports only the distribution based on the state-funded instrument. For comparison, the actual t-value of Medicaid is 56.66 using the state-funded instrument and 73.15 using the Section 1115 waiver instrument.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-9667813/v1/1d0a93c19969c26b757980f2.png"},{"id":109197421,"identity":"c079e0d9-0b50-4b54-9b3a-cd2c0e7087fa","added_by":"auto","created_at":"2026-05-13 13:16:38","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1081630,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMarginal effect of Medicaid with different bandwidths\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNotes: Bandwidth of default kernel function changes from 0.1 to 1. The regression specification is the same as the special regressor model in Table 5.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-9667813/v1/f7738bac8cd4dd731df955a8.png"},{"id":109219654,"identity":"9ed3716f-5669-4fee-b0fc-5b8803a4715d","added_by":"auto","created_at":"2026-05-13 19:58:52","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3285322,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9667813/v1/3a95d33a-d52b-4053-8c2a-1484c073f5b0.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eThe Impact of Medicaid Eligibility on SSI Participation: New Evidence from State-Funded Programs and Section 1115 Waivers\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe Supplemental Security Income (SSI) program occupies a central position in the U.S. safety net. It provides cash assistance to low-income individuals who are aged, blind, or disabled, while also serving as an important gateway to Medicaid in many states (Yelowitz 1998). This institutional linkage means that changes in Medicaid eligibility can reshape the incentives surrounding SSI participation. When Medicaid becomes available outside the SSI system, the value of applying for SSI may decline (Burns and Dague 2017). At the same time, broader Medicaid eligibility may increase contact with medical providers and public assistance agencies, potentially encouraging SSI take-up among individuals who were already close to program eligibility. Using CPS and SIPP data from 2000 to 2013, this paper finds that Medicaid coverage induced by simulated eligibility increases SSI participation by 5.3 to 6.9 percentage points. The result suggests that Medicaid and SSI can operate as complements rather than substitutes, implying that Medicaid expansion has spillovers across safety-net programs.\u003c/p\u003e \u003cp\u003eThis ambiguity motivates the policy question studied in this paper. Medicaid and SSI may operate as substitutes if expanded Medicaid coverage reduces the need to use SSI as a route to public health insurance. They may instead operate as complements if Medicaid eligibility increases awareness of public benefits (Bhargava and Manoli 2015; Finkelstein and Notowidigdo 2019), improves medical documentation, or helps individuals identify disability-related eligibility. The direction of the effect is therefore not determined mechanically by program rules. It depends on how low-income individuals respond to overlapping eligibility systems, application costs (Deshpande and Li 2019), health-care access, and labor-market constraints.\u003c/p\u003e \u003cp\u003eThe pre-ACA period provides a useful setting for studying this interaction. Before the ACA, low-income childless adults had limited access to Medicaid under federal rules, and coverage expansions depended heavily on state policy choices. Some states expanded coverage through state-funded programs, while others used Section 1115 waivers (Burns and Dague 2017; Dague et al. 2017; Bradley and Sabik 2019) to extend eligibility to groups that were otherwise excluded. These policy changes created substantial cross-state and over-time variation in Medicaid availability for childless adults, a group for whom SSI could otherwise be one of the most salient routes to public health insurance.\u003c/p\u003e \u003cp\u003eExisting empirical evidence is mixed. Several studies emphasize a substitution channel (Burns and Dague 2017), finding that Medicaid expansions reduced SSI participation or SSI applications by lowering the need to obtain Medicaid through disability-based programs. Other evidence suggests that the relationship depends strongly on study design (Ne\u0026rsquo;eman and Maestas 2023), population (Staiger et al. 2024), and institutional setting (Baicker et al. 2014; Maestas et al. 2014; Schmidt et al. 2020). This paper adds to this literature by focusing on individual-level Medicaid coverage in a binary SSI participation model and using simulated eligibility to isolate policy-driven variation in Medicaid access.\u003c/p\u003e \u003cp\u003eThe empirical challenge is that Medicaid coverage is not randomly assigned. Individuals who obtain Medicaid may differ systematically from those who do not in health status, labor-market attachment, benefit knowledge, and underlying disability risk. These same unobserved factors may also affect SSI participation. To address this concern, the paper uses simulated Medicaid eligibility measures (Cutler and Gruber 1996; Currie and Gruber 1996) based on state-funded programs and Section 1115 waivers as instruments and applies the special regressor approach (Lewbel 2000; Dong and Lewbel 2015) for binary outcome models with endogenous regressors. This approach complements conventional IV-LPM and IV-Probit specifications by providing a semiparametric framework that relies less on linear probability assumptions.\u003c/p\u003e \u003cp\u003eThe main results indicate a positive relationship between Medicaid coverage and SSI participation. In the preferred special regressor estimates, Medicaid coverage induced by simulated eligibility increases SSI participation by 6.6 and 6.9 percentage points in the CPS under state-funded and Section 1115 waiver programs, respectively. The corresponding estimates in the SIPP are 5.3 and 6.4 percentage points. The effects are substantially larger among unemployed individuals, consistent with the view that Medicaid eligibility is more closely linked to SSI participation for adults with weaker labor-market attachment and lower earnings capacity (Autor and Duggan 2003; Maestas et al. 2013; Garthwaite et al. 2014). Gender heterogeneity is also present in the CPS, although it is less stable in the SIPP.\u003c/p\u003e \u003cp\u003eThe paper makes three contributions. First, it reframes the Medicaid-SSI relationship as an empirical question about substitution versus complementarity across welfare programs. Second, it distinguishes two pre-ACA policy channels-state-funded programs and Section 1115 waivers-that expanded Medicaid access for childless adults. Third, it applies the special regressor approach to a binary-outcome setting in which actual Medicaid coverage is endogenous, thereby complementing the existing DID and IV-LPM evidence.\u003c/p\u003e \u003cp\u003eThe results are therefore best interpreted as evidence on program interaction rather than as a claim that Medicaid mechanically raises SSI participation for all individuals. The findings indicate that, for low-income childless adults affected by Medicaid eligibility expansions, Medicaid access may increase the probability of SSI receipt through welfare-system contact (Bhargava and Manoli 2015), medical documentation (Messel et al. 2023), and latent disability-related eligibility (Deshpande and Li 2019).\u003c/p\u003e \u003cp\u003eThe paper proceeds as follows. Section 2 reviews the literature on Medicaid-SSI interactions and simulated eligibility. Section 3 describes the data, variable construction, and empirical strategy. Section 4 presents the baseline estimates, special regressor results, heterogeneity analyses, and robustness checks. Section 5 concludes by discussing the implications for the design of overlapping welfare programs.\u003c/p\u003e"},{"header":"2. Literature Review","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Previous Research on SSI/SSDI and Medicaid\u003c/h2\u003e \u003cp\u003eA large literature studies interactions among SSI, SSDI, and Medicaid, with particular attention to whether public health insurance access changes incentives to apply for disability-related cash benefits. Early work (Yelowitz 1998) emphasized the value of Medicaid as a component of SSI participation and showed that the Medicaid link could affect the attractiveness of disability programs (Autor and Duggan 2003; Maestas et al. 2013). This perspective implies that expanding Medicaid outside SSI may reduce SSI participation by weakening the health-insurance motive to apply.\u003c/p\u003e \u003cp\u003eBurns and Dague (2017) provide a key benchmark for this substitution view. Using pre-ACA Medicaid expansions for childless adults, they show that Medicaid coverage made available independently of SSI was associated with a decline in SSI participation among non-disabled childless adults aged 21\u0026ndash;64. Staiger et al. (2024) similarly examine Medicaid expansion and SSI take-up using CPS ASEC data and highlight heterogeneity across racial and ethnic groups. These studies are important because they identify reduced-form policy effects of Medicaid expansions at the state level.\u003c/p\u003e \u003cp\u003eOther studies find smaller, statistically insignificant, or more design-sensitive effects. Evidence from the Massachusetts health reform (Maestas et al. 2014), the Oregon Medicaid experiment (Baicker et al. 2014), and broader analyses of Medicaid expansions (Schmidt et al. 2020) suggests that estimates can vary depending on the population, disability category, timing, and empirical design. Ne\u0026rsquo;eman and Maestas (2023), in particular, emphasize that conclusions about Medicaid and disability-program participation are sensitive to study design. This sensitivity leaves room for complementary mechanisms, especially among individuals whose Medicaid coverage changes through eligibility rules and who may face high application costs (Deshpande and Li 2019) or limited knowledge of benefit programs (Bhargava and Manoli 2015).\u003c/p\u003e \u003cp\u003eThis paper builds on that debate by focusing on whether Medicaid eligibility can increase SSI participation among low-income childless adults. This positive channel is conceptually distinct from the standard substitution mechanism. Medicaid eligibility may increase medical contact (Messel et al. 2023), generate better documentation of health limitations, and expose individuals to administrative information about other public programs (Yelowitz 1998). If these forces dominate the reduction in SSI's health-insurance value, Medicaid and SSI may appear as complements for marginally affected individuals.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Simulated Eligibility (SE)\u003c/h2\u003e \u003cp\u003eSimulated eligibility has been widely used to separate policy variation in public insurance rules from individual-level selection into coverage. The central idea is to construct a measure of eligibility that reflects statutory program rules (Cutler and Gruber 1996; Currie and Gruber 1996) rather than individual choices or unobserved health demand. In Medicaid research, this approach is useful because actual Medicaid coverage may respond to income, employment, health status, and program knowledge, all of which may also be related to SSI participation.\u003c/p\u003e \u003cp\u003eBradley and Sabik (2019) use simulated eligibility to study Medicaid expansions and labor supply among childless adults. Their approach summarizes state-year variation in eligibility rules and uses that variation as an instrument for Medicaid coverage. Related work on public health insurance expansions (Cutler and Gruber 1996; Currie and Gruber 1996) similarly uses simulated eligibility to address endogenous population composition and to isolate policy-driven changes in program access.\u003c/p\u003e \u003cp\u003eFollowing this literature, this paper constructs separate simulated eligibility measures for state-funded programs and Section 1115 waivers. This distinction is important because the two policy channels differed in financing, implementation, and eligibility thresholds. Estimating them separately allows the analysis to ask whether the Medicaid-SSI relationship is similar across different pre-ACA expansion mechanisms.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Data and Methodology","content":"\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Data\u003c/h2\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e3.1.1 State Medicaid programs\u003c/h2\u003e \u003cp\u003eIn the United States, Medicaid eligibility before the ACA varied substantially by demographic group and state policy. Pregnant women, children, and some parents could qualify through federal and state eligibility categories, but low-income childless adults were generally less likely to be covered under standard federal rules. For this group, Medicaid access often depended on whether a state implemented a Section 1115 waiver (Burns and Dague 2017; Dague et al. 2017) or created a state-funded program (Bradley and Sabik 2019) outside the standard federal eligibility pathway.\u003c/p\u003e \u003cp\u003eTo capture this variation, the analysis follows Bradley and Sabik (2019) and uses Appendix Table\u0026nbsp;10 in Bradley and Sabik (2019) as the basis for the Medicaid policy rules. The policy information summarizes state-funded programs and Section 1115 waivers from 2000 to 2013, including implementation years and income thresholds expressed as percentages of the federal poverty level (FPL). Because eligibility thresholds often differed between working and non-working adults, the construction distinguishes between these groups when assigning simulated eligibility. Programs with limited benefits are excluded from the simulated eligibility calculation. Hawaii, Alaska, and the District of Columbia are also excluded because of program complexity, small cell sizes, or overlapping eligibility rules that make it difficult to assign a single policy channel.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e3.1.2 March Current Population Survey (CPS)\u003c/h2\u003e \u003cp\u003eThe first individual-level dataset is the March Current Population Survey (CPS), an annual household survey fielded by the U.S. Census Bureau. The CPS provides nationally representative information on demographic characteristics, labor-market status, income, and participation in public programs, including Medicaid and SSI.\u003c/p\u003e \u003cp\u003eThe CPS sample covers 2000\u0026ndash;2013 and is restricted to childless adults aged 20\u0026ndash;64, the population most directly affected by pre-ACA Medicaid expansions for childless adults. Individuals serving in the military and respondents reporting negative income are excluded to maintain consistency in eligibility measurement and sample composition.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e3.1.3 Survey of Income and Program Participation (SIPP)\u003c/h2\u003e \u003cp\u003eThe second dataset is the Survey of Income and Program Participation (SIPP), also administered by the U.S. Census Bureau. The SIPP provides detailed longitudinal information on income, employment, household characteristics, and program participation, making it useful for cross-validating patterns found in the CPS.\u003c/p\u003e \u003cp\u003eThe analysis uses SIPP core data from 2000 to 2013. Because the SIPP records program participation at a monthly frequency, the dataset contains richer within-year information than the CPS. This structure allows the analysis to examine whether the positive Medicaid-SSI relationship is robust in a longitudinal survey setting, while recognizing that CPS and SIPP differ in sampling design and measurement.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e3.1.4 Dependent variable\u003c/h2\u003e \u003cp\u003eThe dependent variable is SSI participation, defined as whether an individual receives Supplemental Security Income. In the CPS, SSI participation is measured using the SSI-YN item. The variable SSI equals one if the respondent reports receiving SSI and zero otherwise.\u003c/p\u003e \u003cp\u003eIn the SIPP, SSI participation is constructed using the variables RCUTYP03 and RCUTYP04, which identify federal and state SSI receipt. The variable SSI equals one if the respondent is covered by any SSI program and zero otherwise.\u003csup\u003e1\u003c/sup\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e3.1.5 Independent variable\u003c/h2\u003e \u003cp\u003eThe key endogenous explanatory variable is Medicaid coverage. In the CPS, Medicaid coverage is measured using the CAID indicator, which records whether the respondent is covered by Medicaid or a state-specific equivalent program. The variable Medicaid equals one if the respondent reports Medicaid coverage and zero otherwise.\u003c/p\u003e \u003cp\u003eIn the SIPP, Medicaid coverage is constructed from ECDMTH, which identifies whether the respondent is covered by Medicaid in a given month. The variable Medicaid equals one when the respondent reports Medicaid coverage in that month and zero otherwise.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section3\"\u003e \u003ch2\u003e3.1.6 Control variable\u003c/h2\u003e \u003cp\u003eThe analysis includes individual and regional controls commonly used in the welfare and health economics literature (Autor and Duggan 2003; Dague et al. 2017; Bradley and Sabik 2019). Individual-level controls include gender, age and age squared, marital status, employment status, educational attainment, and race or ethnicity. Marital status and employment status are measured as binary indicators. Race and ethnicity are captured using indicators for White, Black, and Hispanic respondents.\u003c/p\u003e \u003cp\u003eRegional controls include metropolitan status, the number of community hospitals in the state-year, and the natural logarithm of state GDP.\u003csup\u003e2\u003c/sup\u003e These variables account for local economic conditions, health-care supply, and regional differences that may be correlated with both Medicaid coverage and SSI participation.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Methodology\u003c/h2\u003e \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e \u003ch2\u003e3.2.1 Simulated Eligibility (SE)\u003c/h2\u003e \u003cp\u003eA central empirical concern is the endogeneity of Medicaid coverage. Individuals who obtain Medicaid may differ from non-covered individuals in unobserved health, disability risk, program knowledge, and labor-market attachment. To address this issue, the paper uses simulated eligibility as an instrument (Cutler and Gruber 1996; Currie and Gruber 1996; Bradley and Sabik 2019). Simulated eligibility captures policy-induced variation in Medicaid access generated by state-funded programs and Section 1115 waivers, rather than relying solely on observed Medicaid coverage choices.\u003c/p\u003e \u003cp\u003eTo construct the simulated eligibility variables, the analysis uses childless adults aged 20\u0026ndash;64 from the March CPS between 2000 and 2013. Family income is converted into a percentage of the FPL, and respondents are assigned eligibility according to the relevant state-year program rules in Appendix Table\u0026nbsp;10 in Bradley and Sabik (2019). Because some states operated both state-funded programs and Section 1115 waivers, the paper constructs two separate measures: se_state for state-funded programs and se_1115 for Section 1115 waivers.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section3\"\u003e \u003ch2\u003e3.2.2 Special Regressor Estimator\u003c/h2\u003e \u003cp\u003eIn policy settings with a binary outcome and an endogenous binary explanatory variable, the instrumental-variable linear probability model is often used as a benchmark. However, the LPM may generate fitted probabilities outside the unit interval and can impose restrictive linearity assumptions. The special regressor estimator provides an alternative semiparametric approach for binary-choice models with endogenous regressors, following Lewbel (2000), Dong and Lewbel (2015), and related applications.\u003c/p\u003e \u003cp\u003eThe starting point is a binary-choice model in which the observed outcome equals one when the latent index is positive:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:D=I\\left({\\varvec{X}}^{\\mathbf{{\\prime\\:}}}\\beta\\:+\\epsilon\\:\\ge\\:0\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere D represents the binary outcome variable, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\varvec{X}}^{\\mathbf{{\\prime\\:}}}\\)\u003c/span\u003e\u003c/span\u003e is the vector containing all observable regressors, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\epsilon\\:\\)\u003c/span\u003e\u003c/span\u003e follows a zero-mean distribution, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:I\\left({\\bullet\\:}\\right)\\)\u003c/span\u003e\u003c/span\u003e represents the function that takes value one if the latent variable inside is positive, and zero otherwise. And to get the special regressor model, we need two more variables, an instrumental variable (Z) and a special regressor (V). And the special regressor should not be a part of the instrumental variables.\u003c/p\u003e \u003cp\u003eIn our study, the instrumental variable we choose is SE, for it satisfies the requirements of relevance and exogeneity. And we adopt the variable \u003cem\u003eage\u003c/em\u003e as our special regressor (V), which meets the three criteria (Lewbel 2014). It is continuously distributed with wide support, because \u003cem\u003eage\u003c/em\u003e ranges from 20 to 64. And it is also exogenous, for the sample\u0026rsquo;s age is predetermined with respect to the policy variation used to construct simulated eligibility. In the empirical sample, the special-regressor monotonicity condition requires \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:E(D\\mid\\:\\varvec{X},V)\\)\u003c/span\u003e\u003c/span\u003e to be monotonic in \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:V\\)\u003c/span\u003e\u003c/span\u003e. This condition is plausible because SSI participation generally increases with age, and Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows a positive relationship between age and SSI participation in both datasets. Prior studies show that the probability of receiving welfare benefits such as SSI increases with age. And in both of our two datasets, a fitted kernel-weighted local polynomial regression of SSI participation on age provides evidence that there is a positive relationship between them (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) (Autor and Duggan 2003; Maestas et al. 2013). These three properties make \u003cem\u003eage\u003c/em\u003e a useful special regressor in this model.\u003c/p\u003e \u003cp\u003eFollowing the estimation strategy summarized by Bontemps and Nauges (2016), the special regressor method transforms the binary outcome into a form that can be estimated using instrumental variables. The implementation proceeds in four steps.\u003c/p\u003e \u003cp\u003eFirst, define the auxiliary variable T as follows:\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:T=\\frac{D-I(V\\ge\\:0)}{{f}_{V|Z}\\left(V|Z\\right)}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{f}_{V|Z}\\left(V\\right|Z)\\)\u003c/span\u003e\u003c/span\u003e presents the conditional probability density function of V given Z. Then, based on the assumptions, it could be shown that \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:E\\left(T\\right|Z)=E({\\varvec{X}}^{\\mathbf{{\\prime\\:}}}\\beta\\:+\\epsilon\\:\\left|Z\\right)\\)\u003c/span\u003e\u003c/span\u003e, where \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\epsilon\\:\\)\u003c/span\u003e\u003c/span\u003e is assumed independent of \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:V|Z\\)\u003c/span\u003e\u003c/span\u003e. This implies the following condition that \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:E\\left(Z{\\varvec{X}}^{\\mathbf{{\\prime\\:}}}\\right)\\beta\\:=E\\left(ZT\\right)\\)\u003c/span\u003e\u003c/span\u003e, which leads to the following expression for \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\beta\\:\\)\u003c/span\u003e\u003c/span\u003e.\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e\n$$\\:\\beta\\:={\\left[E\\left(\\varvec{X}{Z}^{{\\prime\\:}}\\right){E\\left(\\varvec{Z}{Z}^{{\\prime\\:}}\\right)}^{-1}E\\left(Z{\\varvec{X}}^{\\mathbf{{\\prime\\:}}}\\right)\\right]}^{-1}\\hspace{0.25em}E\\left(\\varvec{X}{Z}^{{\\prime\\:}}\\right){E\\left(Z{Z}^{{\\prime\\:}}\\right)}^{-1}E\\left(ZT\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhich is the definition of a linear two-stage least squares regression of \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:T\\)\u003c/span\u003e\u003c/span\u003e on \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\varvec{X}\\)\u003c/span\u003e\u003c/span\u003e by using instrumental variable \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:Z\\)\u003c/span\u003e\u003c/span\u003e. To implement this estimator in practice, we proceed with the following steps:\u003c/p\u003e \u003cp\u003eStep 1: The special regressor V is first centered. Then V is regressed on the observable covariates X and the instrument Z using ordinary least squares:\u003cdiv id=\"Equd\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equd\" name=\"EquationSource\"\u003e\n$$\\:{\\widehat{U}}_{i}={V}_{i}-\\left({\\varvec{X}}^{\\mathbf{{\\prime\\:}}}{\\widehat{b}}_{X}+{Z}^{{\\prime\\:}}{\\widehat{b}}_{Z}\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\widehat{b}}_{X}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\widehat{b}}_{Z}\\)\u003c/span\u003e\u003c/span\u003e are OLS estimated coefficients for variables in \u003cb\u003eX\u003c/b\u003e and Z, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\widehat{U}}_{i}\\)\u003c/span\u003e\u003c/span\u003e are the residuals.\u003c/p\u003e \u003cp\u003eStep 2: Using the residuals \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\widehat{U}}_{i}\\)\u003c/span\u003e\u003c/span\u003e, estimate their density via a kernel density estimator, which can be written as\u003cdiv id=\"Eque\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Eque\" name=\"EquationSource\"\u003e\n$$\\:{\\widehat{f}}_{h}\\left(u\\right)=\\frac{1}{nh}\\sum\\:_{j=1}^{n}K\\left(\\frac{{\\widehat{U}}_{j}-u}{h}\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:K\\left({\\bullet\\:}\\right)\\)\u003c/span\u003e\u003c/span\u003e is a kernel function, and \u003cem\u003eh\u003c/em\u003e is the bandwidth parameter. This yields a density estimate \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\widehat{f}}_{i}=\\widehat{f}\\left({\\widehat{U}}_{i}\\right)\\)\u003c/span\u003e\u003c/span\u003e. An alternative density estimator, the sorted-data estimator proposed by Lewbel and Schennach (2007) can also be used.\u003c/p\u003e \u003cp\u003eStep 3: After obtaining the density estimator, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\widehat{T}}_{i}\\)\u003c/span\u003e\u003c/span\u003e for each observation \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:i\\)\u003c/span\u003e\u003c/span\u003e could be calculated as:\u003cdiv id=\"Equf\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equf\" name=\"EquationSource\"\u003e\n$$\\:{\\widehat{T}}_{i}=\\frac{{D}_{i}-I\\left({V}_{i}\\ge\\:0\\right)}{{\\widehat{f}}_{i}}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere the original binary outcome \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{D}_{i}\\)\u003c/span\u003e\u003c/span\u003e is transformed into a continuous form suitable for later instrumental variable regression.\u003c/p\u003e \u003cp\u003eStep 4: Finally, the transformed outcome is estimated using a two-stage least squares regression on the covariates, with the simulated eligibility measure serving as the instrument. The resulting estimate captures the effect of the endogenous Medicaid coverage variable under the special regressor assumptions.\u003c/p\u003e \u003cp\u003eAfter identifying V, it can be added to the binary choice model, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\varvec{X}}^{\\mathbf{{\\prime\\:}}}\\)\u003c/span\u003e\u003c/span\u003e can be divided into endogenous (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\varvec{X}}^{\\varvec{e}}\\)\u003c/span\u003e\u003c/span\u003e) and exogenous (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\varvec{X}}^{0}\\)\u003c/span\u003e\u003c/span\u003e) covariates. The SR model can therefore be written as follows:\u003cdiv id=\"Equg\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equg\" name=\"EquationSource\"\u003e\n$$\\:D=I\\left({\\varvec{X}}^{e}{\\beta\\:}_{e}+{\\varvec{X}}^{0}{\\beta\\:}_{0}+V+\\epsilon\\:\\ge\\:0\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eIn the empirical application, the model is estimated using a standard kernel density estimator. The robustness checks vary the kernel function and bandwidth. The estimating equation can be written as:\u003cdiv id=\"Equh\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equh\" name=\"EquationSource\"\u003e\n$$\\:{SSI}_{ist}=I\\left({{M}_{ist}}^{e}{\\beta\\:}^{{\\prime\\:}}+{\\varvec{X}}_{\\varvec{i}\\varvec{s}\\varvec{t}}\\gamma\\:+V+{\\epsilon\\:}_{ist}\\ge\\:0\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eHere, SSI denotes the binary indicator for SSI participation; Medicaid is the endogenous Medicaid coverage variable for individual i in state s and year t; X is the vector of exogenous controls; and age is the special regressor.\u003c/p\u003e \u003cp\u003eBecause coefficients in special regressor models do not have a direct probability interpretation, the analysis reports marginal effects. Following Dong and Lewbel (2015) and Bontemps and Nauges (2016), the marginal effects summarize how Medicaid coverage changes the probability of SSI participation under the estimated binary-choice model.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"4. Results","content":"\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Descriptives\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e defines the variables used in the empirical analysis. Because some states operated state-funded programs and Section 1115 waivers in the same period, the analysis constructs separate simulated eligibility measures for each policy channel. State GDP is transformed into logarithmic form to reduce the influence of extreme values and to capture differences in local economic conditions more flexibly.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e presents descriptive statistics for the CPS and SIPP samples. In the CPS, the final sample includes 229,978 observations, with 183,728 observations in the Section 1115 waiver sample and 68,477 observations in the state-funded program sample. SSI participation is relatively low, at roughly 3 percent, while Medicaid coverage ranges from 8.75 to 10.16 percent across the two policy samples. These patterns are consistent with the fact that both programs target relatively disadvantaged low-income populations (Burns and Dague 2017; Bradley and Sabik 2019).\u003c/p\u003e \u003cp\u003eThe SIPP sample exhibits a similar structure. Simulated eligibility rates are approximately 21.8 percent for Section 1115 waivers and 21.5 percent for state-funded programs, while SSI participation remains much lower. The comparison between CPS and SIPP provides a useful robustness check because the two surveys differ in design, frequency, and measurement of program participation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Baseline Regression\u003c/h2\u003e \u003cdiv id=\"Sec19\" class=\"Section3\"\u003e \u003ch2\u003e4.2.1 Results from the CPS\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e reports baseline LPM estimates for the CPS. These specifications should be interpreted as descriptive associations between Medicaid coverage and SSI participation before the instrumental-variable and special regressor estimates are introduced. Across specifications, the coefficient on Medicaid remains positive and statistically significant.\u003c/p\u003e \u003cp\u003eIn the preferred specification with controls and state and year fixed effects, Medicaid coverage is associated with a 27.0 percentage-point increase in SSI participation. Given the low baseline rate of SSI receipt in the sample, this is an economically large association. The magnitude reinforces the need to address endogeneity directly, since actual Medicaid coverage is likely correlated with unobserved health (Yelowitz 1998), disability (Autor and Duggan 2003; Maestas et al. 2013), and program participation propensities.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section3\"\u003e \u003ch2\u003e4.2.2 Results from the SIPP\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e reports the corresponding baseline estimates using the SIPP. The pattern is similar to the CPS: Medicaid coverage is positively associated with SSI participation across all four specifications. In the specification with controls and state and year fixed effects, the coefficient is 40.3 percentage points.\u003c/p\u003e \u003cp\u003eThese baseline results establish a strong positive association between Medicaid coverage and SSI participation. Because they do not by themselves resolve the endogeneity of Medicaid coverage, the next section turns to instrumental-variable and special regressor estimates.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Estimation Results\u003c/h2\u003e \u003cp\u003eThis section compares the special regressor estimates with conventional IV-Probit and IV-LPM benchmarks. The comparison is useful because the outcome is binary and the treatment variable, Medicaid coverage, is potentially endogenous.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e reports estimates from IV-Probit, IV-LPM, and the special regressor model (Lewbel 2000; Dong and Lewbel 2015; Bontemps and Nauges 2016). Because IV-Probit and special regressor coefficients are not directly interpretable as probability effects, the discussion focuses on the average marginal effects. The IV-Probit estimates provide a nonlinear benchmark, while the IV-LPM estimates provide a conventional linear probability benchmark. The special regressor estimates are the preferred semiparametric estimates because they address the endogeneity of Medicaid coverage in a binary-outcome framework without relying on the linear probability specification.\u003c/p\u003e \u003cp\u003eIn the CPS, the IV-Probit marginal effects are 7.0 percentage points under the state-funded instrument and 7.4 percentage points under the Section 1115 waiver instrument. These estimates are very close to the special regressor marginal effects, which are 6.6 and 6.9 percentage points, respectively. By contrast, the IV-LPM estimates are substantially larger, at 24.3 and 25.3 percentage points. This pattern suggests that the linear probability framework may overstate the magnitude of the relationship between Medicaid coverage and SSI participation in this binary-outcome setting.\u003c/p\u003e \u003cp\u003eA similar pattern appears in the SIPP. The IV-Probit marginal effects are 7.6 percentage points under the state-funded instrument and 9.1 percentage points under the Section 1115 waiver instrument. The corresponding special regressor marginal effects are 5.3 and 6.4 percentage points. As in the CPS, the IV-LPM estimates are much larger, at 38.4 and 35.4 percentage points. Taken together, the results indicate that Medicaid coverage induced by simulated eligibility is positively associated with SSI participation across both datasets and both policy instruments. The magnitude of the effect is more moderate under the nonlinear and semiparametric specifications than under the linear probability model.\u003c/p\u003e \u003cp\u003eOverall, the estimates in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e support the interpretation that Medicaid and SSI may operate as complements for low-income childless adults in this setting (Yelowitz 1998). Rather than simply reducing the value of SSI as a pathway to Medicaid, Medicaid coverage may increase contact with the public assistance system (Bhargava and Manoli 2015), improve access to medical documentation (Messel et al. 2023), and reveal latent eligibility for disability-related benefits (Deshpande and Li 2019). This interpretation is consistent with the positive and stable marginal effects from the IV-Probit and special regressor estimates, while the larger IV-LPM estimates should be interpreted more cautiously as linear probability benchmarks.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Heterogeneity Analysis: Gender and Employment Status\u003c/h2\u003e \u003cp\u003eBuilding on prior evidence that Medicaid effects vary by demographic group and study design (Staiger et al. 2024; Ne\u0026rsquo;eman and Maestas 2023), the analysis next examines heterogeneity by gender and employment status. The subgroup estimates use the IV-LPM specification from Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e and are reported separately for the CPS and SIPP.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e shows that Medicaid coverage has a positive and statistically significant effect for both men and women. In the CPS, the estimated effects are larger for women than for men under both state-funded programs and Section 1115 waivers. In the SIPP, women experience a larger effect under state-funded programs, while the gender difference is less pronounced under Section 1115 waivers. This pattern suggests that gender heterogeneity is present but should be interpreted cautiously because it is not fully stable across datasets.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e shows a more robust pattern by employment status. In both the CPS and SIPP, the estimated effect is substantially larger among unemployed individuals than among employed individuals. For example, in the CPS state-funded specification, Medicaid coverage increases SSI participation by 30.7 percentage points among unemployed individuals, compared with 9.9 percentage points among employed individuals. The same qualitative pattern appears under Section 1115 waivers and in the SIPP.\u003c/p\u003e \u003cp\u003eThe employment heterogeneity is consistent with the view that Medicaid eligibility is especially relevant for individuals with weaker labor-market attachment and lower earnings capacity (Garthwaite et al. 2014; Dague et al. 2017). For this group, Medicaid coverage may increase interaction with the welfare system and may also coincide with characteristics that make SSI participation more likely. The gender results are more suggestive, whereas the employment-status results are consistent across both datasets and policy channels.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Robustness Checks\u003c/h2\u003e \u003cdiv id=\"Sec24\" class=\"Section3\"\u003e \u003ch2\u003e4.4.1 Placebo Tests\u003c/h2\u003e \u003cp\u003eThe analysis conducts two placebo exercises to examine whether the main results are likely to be driven by broad state-level trends or random assignment of policy exposure. The first placebo test restricts the sample to high-income childless adults, who are unlikely to be affected by Medicaid eligibility expansions targeted at low-income populations.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e shows that the estimated marginal effects for the high-income group are all statistically insignificant. In particular, CPS estimates are close to zero, while SIPP estimates are imprecise and statistically insignificant. This pattern supports the interpretation that the main estimates are concentrated among the intended low-income population rather than reflecting general changes in SSI participation unrelated to Medicaid eligibility.\u003c/p\u003e \u003cp\u003eThe second placebo exercise replaces the true simulated eligibility instruments with randomly generated false eligibility measures over 1,000 simulations. Figures\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e show that the resulting placebo t-statistics are concentrated around zero and rarely exceed the conventional 5 percent critical value, while the actual estimates lie far in the right tail of the placebo distribution. This pattern indicates that the estimated relationship is driven by actual Medicaid eligibility variation rather than random assignment.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003e4.4.3 Sensitivity Tests\u003c/h2\u003e \u003cp\u003eThe analysis also examines the sensitivity of the special regressor estimates (Lewbel and Schennach 2007; Bontemps and Nauges 2016) to the choice of kernel function. Table\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e reports estimates that replace the default Epanechnikov kernel with rectangle and triangle kernels under the default bandwidth. Across both datasets and both simulated eligibility instruments, the marginal effects remain positive and statistically significant, and the magnitudes are close to the preferred estimates.\u003c/p\u003e \u003cp\u003eAs an additional sensitivity check, Table\u0026nbsp;\u003cspan refid=\"Tab11\" class=\"InternalRef\"\u003e11\u003c/span\u003e varies the bandwidth parameter of the default kernel from 0.1 to 1.0. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e plots the resulting marginal effects. The estimates remain positive across all bandwidths. In the CPS, the marginal effects are highly stable. In the SIPP, the state-funded estimates decline modestly as the bandwidth increases, while the Section 1115 estimates remain nearly flat. Overall, the bandwidth exercise indicates that the main conclusion does not depend on a particular smoothing parameter.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThis paper examines whether Medicaid eligibility expansions for low-income childless adults affected SSI participation before the Affordable Care Act. Using CPS and SIPP data from 2000 to 2013, simulated eligibility instruments based on state-funded programs and Section 1115 waivers, and a special regressor estimator for binary outcome models with endogenous regressors, the analysis finds a positive relationship between Medicaid coverage and SSI participation.\u003c/p\u003e \u003cp\u003eThe preferred estimates suggest that Medicaid coverage induced by simulated eligibility increases SSI participation by approximately 5 to 7 percentage points across datasets and policy measures. This finding differs from a simple substitution view (Burns and Dague 2017; Baicker et al. 2014) in which expanding Medicaid outside SSI necessarily reduces SSI participation by lowering the value of SSI as a route to health insurance. Instead, the results suggest that Medicaid and SSI may operate as complements for some low-income childless adults (Yelowitz 1998).\u003c/p\u003e \u003cp\u003eThe heterogeneity results provide additional support for this interpretation. The effect is consistently stronger among unemployed individuals, a group with weaker labor-market attachment and lower earnings capacity (Garthwaite et al. 2014; Dague et al. 2017). The gender pattern is more mixed: women experience larger effects in the CPS and under the SIPP state-funded specification, but the difference is less stable under Section 1115 waivers in the SIPP (Staiger et al. 2024).\u003c/p\u003e \u003cp\u003eMethodologically, the paper complements conventional LPM, Probit, and IV specifications by applying the special regressor approach (Lewbel 2000; Dong and Lewbel 2015; Bontemps and Nauges 2016) to a setting with both a binary outcome and endogenous Medicaid coverage. The special regressor estimates are smaller than the baseline and IV-LPM estimates but remain positive across datasets, instruments, kernels, and bandwidth choices.\u003c/p\u003e \u003cp\u003eSubstantively, the findings highlight the importance of coordination across overlapping welfare programs. Expanding Medicaid eligibility may not only change health-insurance coverage; it may also alter participation in cash assistance programs by increasing contact with medical providers (Messel et al. 2023), improving documentation, and exposing individuals to information about related benefits (Bhargava and Manoli 2015; Finkelstein and Notowidigdo 2019; Deshpande and Li 2019). These channels imply that the fiscal and behavioral consequences of Medicaid expansions can extend beyond the Medicaid program itself.\u003c/p\u003e \u003cp\u003eThe analysis has several limitations. The CPS and SIPP differ in survey design and program-participation measurement, and the data do not directly observe the application process, medical documentation, or caseworker interactions. Future research using administrative application records (Deshpande and Li 2019) or health-care utilization data (Messel et al. 2023) could more directly distinguish the awareness, diagnosis, and labor-supply channels underlying the Medicaid-SSI relationship.\u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eConflict of interest disclosure:\u003c/h2\u003e \u003cp\u003eNo potential conflict of interest was disclosed by the author(s).\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding statement:\u003c/h2\u003e \u003cp\u003eThis research did not receive any specific grant from a funding agency.\u003c/p\u003e\u003ch2\u003eAcknowledgement:\u003c/h2\u003e \u003cp\u003eNone.\u003c/p\u003e\u003ch2\u003eData availability statement:\u003c/h2\u003e \u003cp\u003eThe data in this study can be requested from the corresponding author.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAutor, D.H., Duggan, M.G., 2003. The rise in the disability rolls and the decline in unemployment. The Quarterly Journal of Economics 118, 157\u0026ndash;206. doi:10.1162/00335530360535171.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaicker, K., Finkelstein, A., Song, J., Taubman, S., 2014. The impact of Medicaid on labor market activity and program participation: Evidence from the Oregon health insurance experiment. American Economic Review 104, 322\u0026ndash;328. doi:10.1257/aer.104.5.322.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBhargava, S., Manoli, D., 2015. Psychological frictions and the incomplete take-up of social benefits: Evidence from an IRS field experiment. American Economic Review 105, 3489\u0026ndash;3529. doi:10.1257/aer.20121493.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBontemps, C., Nauges, C., 2016. The impact of perceptions in averting-decision models: An application of the special regressor method to drinking water choices. American Journal of Agricultural Economics 98, 297\u0026ndash;313. doi:10.1093/ajae/aav046.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBradley, C.J., Sabik, L.M., 2019. Medicaid expansions and labor supply among low-income childless adults: Evidence from 2000 to 2013. International Journal of Health Economics and Management 19, 235\u0026ndash;272. doi:10.1007/s10754-018-9248-x.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBurns, M., Dague, L., 2017. The effect of expanding Medicaid eligibility on supplemental security income program participation. Journal of Public Economics 149, 20\u0026ndash;34. doi:10.1016/j.jpubeco.2017.03.004.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCurrie, J., Gruber, J., 1996. Health insurance eligibility, utilization of medical care, and child health. The Quarterly Journal of Economics 111, 431\u0026ndash;466. doi:10.2307/2946684.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCutler, D.M., Gruber, J., 1996. Does public insurance crowd out private insurance? The Quarterly Journal of Economics 111, 391\u0026ndash;430. doi:10.2307/2946683.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDague, L., DeLeire, T., Leininger, L., 2017. The effect of public insurance coverage for childless adults on labor supply. American Economic Journal: Economic Policy 9, 124\u0026ndash;154. doi:10.1257/pol.20150059.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeshpande, M., Li, Y., 2019. Who is screened out? application costs and the targeting of disability programs. American Economic Journal: Economic Policy 11, 213\u0026ndash;248. doi:10.1257/pol.20180076.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDong, Y., Lewbel, A., 2015. A simple estimator for binary choice models with endogenous regressors. Econometric Reviews 34, 82\u0026ndash;105. doi:10.1080/07474938.2014.944470.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFinkelstein, A., Notowidigdo, M.J., 2019. Take-up and targeting: Experimental evidence from SNAP. The Quarterly Journal of Economics 134, 1505\u0026ndash;1556. doi:10.1093/qje/qjz013.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarthwaite, C., Gross, T., Notowidigdo, M.J., 2014. Public health insurance, labor supply, and employment lock. The Quarterly Journal of Economics 129, 653\u0026ndash;696. doi:10.1093/qje/qju005.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLewbel, A., 2000. Semiparametric qualitative response model estimation with unknown heteroscedasticity or instrumental variables. Journal of Econometrics 97, 145\u0026ndash;177. doi:10.1016/S0304-4076(00)00015-4.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLewbel, A., 2014. An overview of the special regressor method, in: Ullah, A., Racine, J.S., Su, L. (Eds.), The Oxford Handbook of Applied Nonparametric and Semiparametric Econometrics and Statistics. Oxford University Press, Oxford, pp. 38\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLewbel, A., Schennach, S.M., 2007. A simple ordered data estimator for inverse density weighted expectations. Journal of Econometrics 136, 189\u0026ndash;211. doi:10.1016/j.jeconom.2005.08.005.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaestas, N., Mullen, K.J., Strand, A., 2013. Does disability insurance receipt discourage work? using examiner assignment to estimate causal effects of SSDI receipt. American Economic Review 103, 1797\u0026ndash;1829. doi:10.1257/aer.103.5.1797.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaestas, N., Mullen, K.J., Strand, A., 2014. Disability insurance and health insurance reform: Evidence from Massachusetts. American Economic Review 104, 329\u0026ndash;335. doi:10.1257/aer.104.5.329.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMessel, M., Swensen, I., Urban, C., 2023. The effects of expanding access to mental health services on SS(D)I applications and awards. Labour Economics 81, 102339. doi:10.1016/j.labeco.2023.102339.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNe'eman, A., Maestas, N., 2023. How Does Medicaid Expansion Impact Income Support Program Participation and Employment for Different Types of People with Disabilities? NBER Working Paper 31816. National Bureau of Economic Research. doi:10.3386/w31816.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSanderson, E., Windmeijer, F., 2016. A weak instrument F-test in linear IV models with multiple endogenous variables. Journal of Econometrics 190, 212\u0026ndash;221. doi:10.1016/j.jeconom.2015.06.004.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmidt, L., Shore-Sheppard, L.D., Watson, T., 2020. The impact of the ACA Medicaid expansion on disability program applications. American Journal of Health Economics 6, 444\u0026ndash;476. doi:10.1086/710525.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStaiger, B., Helfer, M.S., Van Parys, J., 2024. The effect of Medicaid expansion on the take-up of disability benefits by race and ethnicity. Health Economics 33, 526\u0026ndash;540. doi:10.1002/hec.4783.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYelowitz, A.S., 1998. Why did the SSI-disabled program grow so much? disentangling the effect of Medicaid. Journal of Health Economics 17, 321\u0026ndash;349. doi:10.1016/S0167-6296(97)00024-6.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Footnotes","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003e In the SIPP, SSI receipt can be separately constructed for the policy-channel specifications. Therefore, the SIPP regressions use the SSI measure corresponding to the Section 1115 waiver and state-funded program specifications, respectively. Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e reports these SIPP SSI measures separately, while the CPS reports a single SSI participation measure because the CPS does not provide the same separate SSI classification.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e The number of community hospitals is obtained from the 1999\u0026ndash;2024 AHA Annual Survey. State GDP is from the U.S. Bureau of Economic Analysis SASUMMARY state annual summary statistics.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":" \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eVariable Definitions\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDefinition\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWhether the respondent receives Supplemental Security Income\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMedicaid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWhether the respondent is covered by Medicaid\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSE_state\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWhether the respondent is eligible to receive Medicaid coverage through state-funded programs\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSE_1115\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWhether the respondent is eligible to receive Medicaid coverage through Section 1115 waiver programs\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eThe age of the respondent\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWhether the respondent\u0026rsquo;s gender of birth is female\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHighEdu\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWhether the respondent has attained a degree higher than high school\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWork\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWhether the respondent is currently employed\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMarried\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWhether the respondent is currently married\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWhether the respondent currently lives in a metropolitan area\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003elnGDP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNatural logarithm of state GDP\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospital\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNumber of community hospitals in the state-year\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWhite\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWhether the respondent belongs to the White racial category\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBlack\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWhether the respondent belongs to the Black racial category\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHispanic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWhether the respondent belongs to the Hispanic racial category\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"2\"\u003e\u003cem\u003eNotes\u003c/em\u003e: This table reports the definitions of the main variables used in the empirical analysis. For binary variables, a value of 1 indicates that the condition described in the definition is satisfied, and a value of 0 indicates otherwise. SSI refers to Supplemental Security Income participation, and Medicaid refers to Medicaid coverage.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescriptive Statistics\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e\u003cb\u003eVariable\u003c/b\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eSIPP\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eCPS\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSE (1115)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSE (state)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSE (1115)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSE (state)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(21.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(21.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(19.44%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(19.02%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSI*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.94%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSI (1115)*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSI (State)*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.79%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.56%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMedicaid*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8.75%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e44\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e51%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e51%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e49%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e49%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMarried*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e47%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e49%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e50%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMSA*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e81%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHighEdu*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e66%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e66%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e59%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWork*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e75%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e77%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e77%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e80%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospital\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e162\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e116\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e160\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e115\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003elnGDP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRace\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBlack*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9.86%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7.47%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWhite*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e82%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e79.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e84.25%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHispanic*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e10.82%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of Obs.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e630724\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e385293\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e183728\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e68477\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003eNotes\u003c/em\u003e: Variables marked with * are binary variables. The table reports means for continuous variables and the proportion equal to 1 for binary variables. Columns labeled SE (1115) and SE (state) refer to the samples used for the Section 1115 waiver and state-funded program analyses, respectively. SSI and Medicaid are measured as participation or coverage indicators. The number of observations varies across columns because the state-funded and Section 1115 waiver samples are defined by different policy variation.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCPS Baseline Regression Results\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(4)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMedicaid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.295***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.268***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.297***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.270***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.001***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.001***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.003***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.003***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHighEdu\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.008***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.007***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWork\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.052***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.052***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMarried\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.017***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.017***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.003***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.002*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospital\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.000***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003elnGDP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.001**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.017\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.012)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWhite\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.005***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.004***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBlack\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.009***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.009***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHispanic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.008***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.005***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.001***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.037***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.001***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.249*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.007)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.147)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e230,314\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e229,978\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e230,314\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e229,978\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR-squared\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.264\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.285\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.266\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.287\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003eNotes\u003c/em\u003e: The dependent variable is SSI participation. Column (1) includes neither control variables nor fixed effects. Column (2) includes individual- and state-level controls. Column (3) includes state and year fixed effects. Column (4) includes both the full set of controls and state and year fixed effects and is the preferred baseline specification. Robust standard errors are in parentheses. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}}p\u0026lt;0.10\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}}p\u0026lt;0.05\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}\\text{*}}p\u0026lt;0.01\\)\u003c/span\u003e\u003c/span\u003e.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSIPP Baseline Regression Results\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(4)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMedicaid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.403***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.402***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.405***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.403***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.001***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.001***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.008***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.008***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHighEdu\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.017***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.017***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWork\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.005***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.004***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMarried\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.006***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.006***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.046***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.046***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospital\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.000***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003elnGDP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.004***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.005)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWhite\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.005***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.006***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBlack\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.002***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.003***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHispanic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.012***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.012***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.002***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.002***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.011\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.062)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1,756,294\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1,581,626\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1,756,294\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1,581,626\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR-squared\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.363\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.404\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.364\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.405\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003eNotes\u003c/em\u003e: The dependent variable is SSI participation. Column (1) includes neither control variables nor fixed effects. Column (2) includes individual- and state-level controls. Column (3) includes state and year fixed effects. Column (4) includes both the full set of controls and state and year fixed effects and is the preferred baseline specification. Robust standard errors are in parentheses. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}}p\u0026lt;0.10\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}}p\u0026lt;0.05\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}\\text{*}}p\u0026lt;0.01\\)\u003c/span\u003e\u003c/span\u003e.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eEstimated Marginal Effects\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEstimation method\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCoefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAverage marginal effect\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCoefficient SE\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePanel A: CPS\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV - Probit (state)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.480***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.070\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.319\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV - Probit (1115)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.897***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.074\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.177\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV - LPM (state)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.243***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.243\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.017\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV - LPM (1115)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.253***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.253\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.010\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKdens (state)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e44.54***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.066\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.023\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKdens (1115)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e41.19***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.069\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.454\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePanel B: SIPP\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV - Probit (state)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.271***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.076\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.110\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV - Probit (1115)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.441***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.091\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.071\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV - LPM (state)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.384***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.384\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV - LPM (1115)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.354***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.354\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKdens (state)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e112.6***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.053\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17.50\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKdens (1115)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e127.9***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.064\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e28.25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003e\u003cem\u003eNotes\u003c/em\u003e: IV-Probit coefficients are reported on the latent-index scale and are not directly interpretable as probability effects. For IV-LPM, the coefficient is directly interpretable as a probability effect. For the special regressor estimates, Kdens denotes the estimator using the default Epanechnikov kernel density method; its coefficients are not directly interpretable, so average marginal effects are reported separately. The reported standard errors refer to the estimated coefficients, and significance stars are based on coefficient estimates and their corresponding standard errors. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}}p\u0026lt;0.10\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}}p\u0026lt;0.05\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}\\text{*}}p\u0026lt;0.01\\)\u003c/span\u003e\u003c/span\u003e.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eFirst-Stage Regression Results of IV-LPM\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e\u003cb\u003eVariables\u003c/b\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eCPS\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eSIPP\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003estate\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1115\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003estate\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1115\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSE_state\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.126***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.137***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.004)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSE_1115\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.130***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.147***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e68,477\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e183,728\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e385,293\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e630,724\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eYES\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSW F\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e999.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2732.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7424.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14557.68\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSW Chi-sq\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1000.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2733.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7424.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14558.62\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003eNotes\u003c/em\u003e: The dependent variable is Medicaid coverage. SE_state and SE_1115 are the excluded instruments used in the IV-LPM specifications. The Sanderson-Windmeijer first-stage chi-squared statistic (SW Chi-sq) and F statistic (SW F) report underidentification and weak-identification diagnostics for the endogenous Medicaid variable (Sanderson and Windmeijer 2016). All specifications include state and year fixed effects. Robust standard errors are in parentheses. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}}p\u0026lt;0.10\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}}p\u0026lt;0.05\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}\\text{*}}p\u0026lt;0.01\\)\u003c/span\u003e\u003c/span\u003e.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSubgroup Analysis: Gender\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c3\" namest=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePanel A: CPS\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e34,792\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e33,685\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState and Year FEs with CV (state)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.216***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.258***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.029)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.021)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e93,116\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e90,612\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState and Year FEs with CV (1115)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.237***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.260***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.013)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c3\" namest=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePanel B: SIPP\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e189,873\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e195,420\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState and Year FEs with CV (state)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.331***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.407***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.014)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.008)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e308,202\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e322,522\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState and Year FEs with CV (1115)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.365***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.351***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.009)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.006)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003e\u003cem\u003eNotes\u003c/em\u003e: Each coefficient is estimated from a separate IV-LPM regression using the specification in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e. The sample is split by gender as indicated in the column header. The state-funded and Section 1115 waiver estimates are reported separately for the CPS and SIPP samples. Robust standard errors are in parentheses. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}}p\u0026lt;0.10\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}}p\u0026lt;0.05\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}\\text{*}}p\u0026lt;0.01\\)\u003c/span\u003e\u003c/span\u003e.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSubgroup Analysis: Employment Status\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUnemployed\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eEmployed\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c3\" namest=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePanel A: CPS\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13,682\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e54,795\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState and Year FEs with CV (state)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.307***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.099***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.036)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.015)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42,352\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e141,878\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState and Year FEs with CV (1115)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.343***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.086***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.019)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.009)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c3\" namest=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePanel B: SIPP\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e88,451\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e296,842\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState and Year FEs with CV (state)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.434***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.199***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.012)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.009)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e160,185\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e470,539\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState and Year FEs with CV (1115)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.486***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.068***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.008)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.005)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003e\u003cem\u003eNotes\u003c/em\u003e: Each coefficient is estimated from a separate IV-LPM regression using the specification in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e. The sample is split by employment status as indicated in the column header. The state-funded and Section 1115 waiver estimates are reported separately for the CPS and SIPP samples. Robust standard errors are in parentheses. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}}p\u0026lt;0.10\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}}p\u0026lt;0.05\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}\\text{*}}p\u0026lt;0.01\\)\u003c/span\u003e\u003c/span\u003e.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab9\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 9\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePlacebo Test: High-Income Group\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCoefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMarginal effect\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eStandard Error\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePanel A: High-income group in CPS\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState-funded\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-26.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e51.672\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSection 1115 waiver\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-1.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e50.931\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePanel B: High-income group in SIPP\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState-funded\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e211.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e132.1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSection 1115 waiver\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-2306.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.099\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1891.2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003e\u003cem\u003eNotes\u003c/em\u003e: The high-income group is defined as individuals with total family income at or above the 60th percentile within each sample year. The table reports placebo estimates from the special regressor model. Marginal effects are average marginal effects on SSI participation. Reported standard errors refer to the estimated coefficients. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}}p\u0026lt;0.10\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}}p\u0026lt;0.05\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}\\text{*}}p\u0026lt;0.01\\)\u003c/span\u003e\u003c/span\u003e.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab10\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 10\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSpecial Regressor Model Results with Different Kernel Functions\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eData\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eType of kernel function\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAverage marginal effect\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCoefficient SE\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePanel A: CPS\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState-funded\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003erectangle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.066***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.426\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003etriangle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.394\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSection 1115 waiver\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003erectangle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.062***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.447\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003etriangle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.063***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.435\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePanel B: SIPP\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eState-funded\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003erectangle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.072***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21.71\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003etriangle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21.74\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSection 1115 waiver\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003erectangle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.066***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e28.44\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003etriangle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.065***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e28.71\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003e\u003cem\u003eNotes\u003c/em\u003e: Rectangle and triangle are alternative kernel functions used in the special regressor model. Each regression uses the Kdens specification in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e with a different kernel function. Marginal effects are average marginal effects on SSI participation. Reported standard errors refer to the estimated coefficients. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}}p\u0026lt;0.10\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}}p\u0026lt;0.05\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}\\text{*}}p\u0026lt;0.01\\)\u003c/span\u003e\u003c/span\u003e.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab11\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 11\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSpecial Regressor Model Results with Different Bandwidths\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eData\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBandwidth\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eState-funded ME\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eState-funded SE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSection 1115 ME\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSection 1115 SE\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCPS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.287\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.061***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.407\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.065***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.400\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.061***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.418\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.409\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.062***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.448\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.395\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.062***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.439\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.401\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.062***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.437\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.400\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.062***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.440\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.399\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.062***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.437\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.401\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.063***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.437\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.399\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.063***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.439\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.397\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.062***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.437\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSIPP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.079***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e26.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.063***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e28.95\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.075***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e23.85\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e30.44\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.072***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.065***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e28.38\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.068***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.066***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e29.23\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.066***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.066***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e29.26\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.062***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.065***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e28.73\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.063***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.065***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e28.60\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.065***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e28.81\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.064***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.065***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e29.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.063***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.065***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e28.88\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"6\"\u003e\u003cem\u003eNotes\u003c/em\u003e: The special regressor model uses the default Epanechnikov kernel. Each specification uses a different bandwidth from 0.1 to 1.0. The bandwidth controls the degree of smoothing in the kernel density estimation: smaller bandwidths place more weight on nearby observations, while larger bandwidths produce smoother estimates. Marginal effects are average marginal effects on SSI participation. Reported standard errors refer to the estimated coefficients. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}}p\u0026lt;0.10\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}}p\u0026lt;0.05\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{}^{\\text{*}\\text{*}\\text{*}}p\u0026lt;0.01\\)\u003c/span\u003e\u003c/span\u003e.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e "}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"University of Colorado Denver","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Medicaid, Supplemental Security Income, simulated eligibility, special regressor estimator, welfare program interactions","lastPublishedDoi":"10.21203/rs.3.rs-9667813/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9667813/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis paper studies whether Medicaid expansions for low-income childless adults changed participation in Supplemental Security Income (SSI) before the Affordable Care Act. The expected direction of this relationship is theoretically ambiguous. Medicaid made available outside SSI can weaken the incentive to use SSI as a pathway to public insurance, but it can also increase SSI take-up by expanding contact with health providers and the welfare system, improving medical documentation, and revealing latent disability-related eligibility. Using data from the Current Population Survey and the Survey of Income and Program Participation from 2000 to 2013, we construct simulated Medicaid eligibility measures for state-funded programs and Section 1115 waivers and estimate the relationship using IV benchmarks and a special regressor estimator for binary outcome models with endogenous Medicaid coverage. The preferred special-regressor estimates indicate that Medicaid coverage induced by simulated eligibility increases SSI participation by 6.6\u0026ndash;6.9 percentage points in the CPS and 5.3\u0026ndash;6.4 percentage points in the SIPP. Effects are larger among unemployed individuals and are suggestively stronger among women in the CPS. The findings imply that Medicaid and SSI can operate as complements for economically vulnerable adults and show how expansions in one safety-net program may reshape participation in another.\u003c/p\u003e","manuscriptTitle":"The Impact of Medicaid Eligibility on SSI Participation: New Evidence from State-Funded Programs and Section 1115 Waivers","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-13 13:16:32","doi":"10.21203/rs.3.rs-9667813/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"433e12bb-0517-42ff-b1d1-729e594ad85b","owner":[],"postedDate":"May 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":67860544,"name":"Health Economics \u0026 Outcomes Research"}],"tags":[],"updatedAt":"2026-05-13T13:16:32+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-13 13:16:32","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9667813","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9667813","identity":"rs-9667813","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}