Time Aggregation and State Dependence in Welfare Receipt

In this paper, we study the dynamics of social assistance benefit receipt in Germany using annual survey data from the German Socio-Economic Panel (SOEP) for the years 1995-2011. Rates of benefit receipt were stable in Germany at around 8% in the 1990s but started rising in 2001 to peak at over 12% in 2006. We show that this increase in the receipt rate can be attributed to lower exit rates from benefit receipt since 2001. In the econometric part of the paper, we study state dependence in social assistance benefit receipt, i.e. the question to what extent benefit receipt today predicts the probability of future benefit receipt. We estimate a series of dynamic random-effects probit models that control for unobserved heterogeneity and endogeneity of the initial conditions and indeed find considerable evidence of state dependence. Our estimates suggest that benefit receipt one year ago raises the likelihood of benefit receipt today by a factor of 3, which corresponds to an average partial effect of about 14 percentage points. The level of state dependence differs between subsamples and is larger in absolute terms for women, Eastern German residents, and migrants, for whom receipt rates are higher. Studying variations over time, we find a rise in the level of state dependence after the 2005 Hartz reforms in Eastern Germany. We attribute this effect to a drop in average predicted entry rates into benefit receipt without a corresponding fall in predicted benefit persistence rates. We do not identify any comparable change in structural dependence for Germany as a whole or other subsamples we look at. Since a reform that contributes to keeping individuals off benefits while doing little to raise exit rates would increase state dependence but might nonetheless be considered as beneficial, our findings should not be understood as a verdict on the success or failure of the Hartz reforms.

In many advanced welfare states, welfare recipients often receive benefits for long periods. This persistence of welfare receipt can be caused by two distinct mechanisms: genuine or spurious state dependence. Knowledge of which of the two mechanisms drives the observed state dependence is important because the policy implications are different. Most of the empirical evidence on state dependence relies on survey data. However, survey data on welfare receipt are subject to substantial measurement error (i.e., misreporting of welfare benefit receipt), which may also bias state dependence estimates. This paper uses rich linked survey and administrative data to measure the effect of misreporting in the survey data on the estimated state dependence in welfare receipt in Germany. We find a rate of underreporting of welfare benefits of 8.6%. Recipients with relatively good labour market chances tend to underreport benefits more frequently. Overreporting benefits is less pronounced with a rate of 1.6%. Within the survey data, we observe more transitions into and out of the welfare system. However, our estimates of state dependence in welfare receipt based on a dynamic random effects model reveal that the effect of misreporting on estimated state dependence is small, even when we distinguish between working and non-working recipients in the model.

PurposeThis paper investigates the employment dynamics of Greek married women.Design/methodology/approachLongitudinal/panel data for the period 1995‐2001 and dynamic discrete choice models are used for estimation purposes.FindingsIt is found that the probability of being employed is influenced by observed individual characteristics (e.g. human capital, fertility and unearned income), while genuine state dependence and unobserved heterogeneity constitute major sources of observed serial persistence. The results show that lagged employment affects current employment decisions in a systematic way, and that the non‐contemporaneous effects of unearned income and fertility correlate with unobserved heterogeneity. The estimated average partial effects reveal that an employed woman in t−1 has a probability of being employed in t that is almost 50 percentage points higher than for a non‐employed woman.Practical implicationsThe presence of state dependence and unobserved heterogeneity imply that the bridging of the female employment gap between Greece and its EU partners is expected to follow a slow, long‐term course.Originality/valueGreek female labour force participation has been studied only under a static analytical framework. This is the first study to investigate employment decisions of Greek married women in an inter‐temporal setting.

The Dynamics of Disability and Benefit Receipt in Britain

Two key issues in the literature on female labor supply are (i) whether persistence in employment status is due to unobserved heterogeneity or state dependence, and (ii) whether fertility is exogenous to labor supply. Until recently, the consensus was that unobserved heterogeneity is very important and fertility is endogenous. Hyslop (1999) challenged this. Using a dynamic panel probit model of female labor supply including heterogeneity and state dependence, he found that adding autoregressive errors led to a substantial diminution in the importance of heterogeneity. This, in turn, meant he could not reject that fertility is exogenous. Here, we extend Hyslop (1999) to allow classification error in employment status, using an estimation procedure developed by Keane and Wolpin (2001) and Keane and Sauer (2005). We find that a fairly small amount of classification error is enough to overturn Hyslop's conclusions, leading to overwhelming rejection of the hypothesis of exogenous fertility. © 2009 The Econometric Society.

Are Recipients of Social Assistance 'Benefit Dependent'? Concepts, Measurement and Results for Selected Countries

This paper measures the causal effect of time out of the labor force on subsequent employment of Social Security Disability Insurance (SSDI) applicants and distinguishes it from the discouragement effect of receiving disability benefits. Using a unique Social Security Administration workload database to identify exogenous variation in decision times induced by differences in processing speed among disability examiners to whom applicants are randomly assigned, we find that longer processing times reduce the employment and earnings of SSDI applicants for multiple years following application, with the effects concentrated among applicants awarded benefits during their initial application. A one standard deviation (2.1 month) increase in initial processing time reduces long-run "substantial gainful activity" rates by 0.36 percentage points (3.5%) and long-run annual earnings by $178 (5.1%). Because applicants initially denied benefits spend on average more than 15 additional months appealing their denials, previous estimates of the benefit receipt effect are confounded with the effect of delays on subsequent employment. Accounting separately for these channels, we find that the receipt effect is at least 50% larger than previously estimated. Combining the delay and benefits receipt channels reveals that the SSDI application process reduces subsequent employment of applicants on the margin of award by twice as much as prior literature suggests.

We present the first causal estimates of the effect of Social Security Disability Insurance benefit receipt on labor supply using all program applicants. We use administrative data to match applications to disability examiners and exploit variation in examiners' allowance rates as an instrument for benefit receipt. We find that among the estimated 23 percent of applicants on the margin of program entry, employment would have been 28 percentage points higher had they not received benefits. The effect is heterogeneous, ranging from no effect for those with more severe impairments to 50 percentage points for entrants with relatively less severe impairments. (JEL H55, J14, J22)

Summary We consider dynamic discrete choice models with heterogeneity in both the levels parameter and the state dependence parameter. We first present an empirical analysis that motivates the theoretical analysis which follows. The theoretical analysis considers a simple two-state, first-order Markov chain model without covariates in which both transition probabilities are heterogeneous. Using such a model we are able to derive exact small sample results for bias and mean squared error (MSE). We discuss the maximum likelihood approach and derive two novel estimators. The first is a bias corrected version of the Maximum Likelihood Estimator (MLE) although the second, which we term MIMSE, minimizes the integrated mean square error. The MIMSE estimator is always well defined, has a closed-form expression and inherits the desirable large sample properties of the MLE. Our main finding is that in almost all short panel contexts the MIMSE significantly outperforms the other two estimators in terms of MSE. A final section extends the MIMSE estimator to allow for exogenous covariates.

A Non-Bayesian Theory of State-Dependent Utility

An intensive lifestyle intervention (ILI) has been shown to improve diabetes management and physical function. These benefits could lead to better labor market outcomes, but this has not been previously studied. To estimate the association of an ILI for weight loss in type 2 diabetes with employment, earnings, and disability benefit receipt during and after the intervention. This cohort study included participants with type 2 diabetes and overweight or obesity and compared an ILI with a control condition of diabetes support and education. Data for the original trial were accrued from August 22, 2001, to September 14, 2012. Trial data were linked with Social Security Administration records to investigate whether, relative to the control group, the ILI was associated with improvements in labor market outcomes during and after the intervention period. Difference-in-differences models estimating relative changes in employment, earnings, and disability benefit receipt between the ILI and control groups were used, accounting for prerandomization differences in outcomes for linked participants. Outcome data were analyzed from July 13, 2020, to May 17, 2023. The ILI consisted of sessions with lifestyle counselors, dieticians, exercise specialists, and behavioral therapists on a weekly basis in the first 6 months, decreasing to a monthly basis by the fourth year, designed to achieve and maintain at least 7% weight loss. The control group received group-based diabetes education sessions 3 times annually during the first 4 years, with 1 annual session thereafter. Employment and receipt of federal disability benefits (Supplemental Security Income and Social Security Disability Insurance), earnings, and disability benefit payments from 1994 through 2018. A total of 3091 trial participants were linked with Social Security Administration data (60.1% of 5145 participants initially randomized and 97.0% of 3188 of participants consenting to linkage). Among the 3091 with fully linked data, 1836 (59.4%) were women, and mean (SD) age was 58.4 (6.5) years. Baseline clinical and demographic characteristics were similar between linked participants in the ILI and control groups. Employment increased by 2.9 (95% CI, 0.3-5.5) percentage points for the ILI group relative to controls (P = .03) with no significant relative change in disability benefit receipt (-0.9 [95% CI, -2.1 to 0.3] percentage points; P = .13). The findings of this cohort study suggest that an ILI to prevent the progression and complications of type 2 diabetes was associated with higher levels of employment. Labor market productivity should be considered when evaluating interventions to manage chronic diseases.

The extent to which menopause status contributes to depressive symptoms remains controversial. This study aimed to examine associations between menopausal status and depressive symptoms and the heterogeneity in these associations. We conducted three consecutive national surveys of community-dwelling Australian women aged 18-79 years between October 2013 and July 2017. Depressive symptoms were measured by Beck Depression Inventory II (BDI-II) score (range 0-63). Symptoms were classified into a binary outcome of minimal to mild (BDI-II score <20) and moderate to severe (MS) (BDI-II score ≥20). The average effect (average treatment effect [ATE]) and conditional average effect of the menopausal status on depressive symptoms were investigated by causal forest method. The prevalence of MS depressive symptoms in 10,351 participants was 23.7% (95% CI: 22.9-24.5). Compared with premenopause, the BDI-II score was higher for the menopause transition (ATE = 2.43 units, 95% CI: 1.20-3.65) and postmenopause (ATE = 3.03 units, 95% CI: 1.25-4.81). Compared with premenopause, the menopause transition was associated with an average increase in the prevalence of MS depressive symptoms by 6 percentage points (ATE = 0.06, 95% CI: 0.01-0.12). There were no differences in depressive symptoms between the menopause transition and postmenopause.In menopause transition, the BDI-II score was -1.22 units lower and the prevalence of MS depressive symptoms was 5 percentage points lower among employed women compared with unemployed women. In postmenopause, the BDI-II score was 1.75 units higher for unpartnered women compared with partnered women. The menopause transition and postmenopause are associated with a higher BDI-II score and an increased prevalence of MS depressive symptoms compared with premenopause. These associations are favorably modified by paid employment in the menopause transition and unfavorably by being unpartnered postmenopause.

Impact of a community-policing initiative promoting substance use disorder treatment over criminal charges on arrest recidivism

We seek evidence for economic and social mechanisms that aim to explain the relationship between employment and crime. We use the distinctive features of social welfare for identification. We consider a sample of disadvantaged males from The Netherlands who are observed between ages 18 and 32 on a monthly time scale. We simultaneously model the offending, employment and social welfare variables using a dynamic discrete choice model, where we allow for state dependence, reciprocal effects and time-varying unobserved heterogeneity. We find significant negative bi-directional structural effects between employment and property crime. Robustness checks show that only regular employment is able to significantly reduce the offending probability. Further, a significant uni-directional effect is found for the public assistance category of social welfare on property offending. The results highlight the importance of economic incentives for explaining the relationship between employment and crime for disadvantaged individuals. For these individuals the crime reducing effects from the public assistance category of social welfare are statistically equivalent to those from employment, which suggests the importance of financial gains. Further, the results suggest that stigmatizing effects from offending severely reduce future employment probabilities.

Despite their low cost, the use of email invitations to distribute surveys to medical practitioners have been associated with lower response rates. This research compares the difference in response rates from using email approach plus online completion rather than a mailed invitation letter plus a choice of online or paper completion. A parallel randomised controlled trial was conducted during the 11th annual wave of the nationally representative Medicine in Australia: Balancing Employment and Life (MABEL) longitudinal survey of doctors. The control group was invited using a mailed paper letter (including a paper survey plus instructions to complete online) and three mailed paper reminders. The intervention group was approached in the same way apart from the second reminder when they were approached by email only. The primary outcome is the response rate and the statistical analysis was blinded. 18,247 doctors were randomly allocated to the control (9,125) or intervention group (9,127), with 9,108 and 9,107 included in the analysis. Using intention to treat analysis, the response rate in the intervention group was 35.92% compared to 37.59% in the control group, a difference of -1.66 percentage points (95% CI: -3.06 to -0.26). The difference was larger for General Practitioners (-2.76 percentage points, 95% CI: -4.65 to -0.87) compared to other specialists (-0.47 percentage points, 95% CI: -2.53 to 1.60). For those who supplied an email address, the average treatment effect on the treated was higher at -2.63 percentage points (95% CI: -4.50 to -0.75) for all physicians, -3.17 percentage points (95% CI: -5.83 to -0.53) for General Practitioners, and -2.1 percentage points (95% CI: -4.75 to 0.56) for other specialists. For qualified physicians, using email to invite participants to complete a survey leads to lower response rates compared to a mailed letter. Lower response rates need to be traded off with the lower costs of using email rather than mailed letters.