Search

This paper demonstrates that popular linear fixed-effects panel-data estimators are biased and inconsistent when applied in a discrete-time hazard setting - that is, one in which the outcome variable is a binary dummy indicating an absorbing state, even if the data-generating process is fully consistent with the linear discrete-time hazard model. In addition to conventional survival bias, these estimators suffer from another source of - frequently severe - bias that originates from the data transformation itself and, unlike survival bias, is present even in the absence of any unobserved heterogeneity. We suggest an alternative estimation strategy, which is instrumental variables estimation using first-differences of the exogenous variables as instruments for their levels. Monte Carlo simulations and an empirical application substantiate our theoretical results. May 2021

From an economic perspective, marriage and long-term partnership can be seen as a risk-pooling device. This informal insurance contract is, however, not fully enforceable. Each partner is free to leave when his or her support is needed in case of an adverse life event. An adverse health shock is a prominent example for such events. Since relationship breakdown itself is an extremely stressful experience, partnership may backfire as informal insurance against health risks, if health shocks increase the likelihood of relationship breakdown. We address this question empirically, using survey data from Germany. Results from various matching estimators indicate that adverse shocks to mental health substantially increase the probability of a couple splitting up over the following two years. In contrast, there is little effect of a sharp decrease in physical health on relationship stability. If at all, physical health shocks that hit both partners simultaneously stabilize a relationship.

This paper examines weight loss and the formation of healthy habits through cash rewards in the context of a multi-phase randomized controlled trial involving 700 obese individuals. We find effects of monetary incentives for weight loss of up to EUR 300 on body weight during all experimental phases, including a period of a year and a half following the exposure to the financial rewards. We also find effects on healthy behavior during this follow-up phase. After the initial incentive period, we additionally provided participants who had lost a substantial amount of weight with monetary rewards of up to EUR 500. These had only short-term effects on body weight and healthy behavior. We argue that our findings are best explained by monetarily incentivized participants having formed healthy habits by the time the experiment ended and that only the speed of the transition to the new (health) equilibrium was affected by the additional rewards. Contrary to the pessimistic perspective presented in earlier empirical research on habit formation, our results suggest that a simple program relying on weight loss rewards can result in long-term health behavioral change. Die vorliegende Arbeit untersucht den Zusammenhang von Geldprämien und der Herausbildung gesundheitsförderlicher Verhaltensmuster im Rahmen eines mehrphasigen randomisierten Experiments mit 700 adipösen Teilnehmern. Geldprämien von bis zu 300 € für erfolgreiches Abnehmen zeigen in allen Phasen des Experiments Effekte auf das Körpergewicht der Teilnehmer, einschließlich einer Follow-Up-Phase von eineinhalb Jahren nach Ende der monetären Incentivierung. In der Follow-Up-Phase zeigen sich auch Effekte auf gesundheitsrelevantes Verhalten. Nach Ende der ersten Phase des Experiments, in der die Gewichtsreduktion monetär incentiviert wurde, wurden Teilnehmer die substantiell abgenommen hatten erneut mit Geldprämien von bis zu 500 € incentiviert. Diese zusätzlichen Anreize zeigten lediglich kurzfriste Effekte auf Körpergewicht und gesundheitsrelevantes Verhalten. Diese Asymmetrie in den langfristigen Effekten der beiden Anreize lässt sich am besten dadurch erklären, dass die incentivierten Teilnehmer innerhalb der Dauer des Experiments gesundheitsförderliche Verhaltensroutinen herausgebildet hatten, und dass der das Erreichen dieses neuen ('gesünderen') Gleichgewichts durch den zweiten Anreiz lediglich beschleunigt wurde. Im Gegensatz zur überwiegend pessimistischen Sicht in der Literatur, hinsichtlich der Möglichkeit, mittels temporärer Geldanreize langfristige gesundheitsförderliche Verhaltensänderungen zu erreichen, deuten die Ergebnisse darauf hin, dass temporäre monetäre Abnehmanreize dazu in der Lage sind.

When analyzing what determines the efficiency of production, regressing efficiency scores estimated by DEA on explanatory variables has much intuitive appeal. Simar and Wilson (2007) show that this na¨ıve two-stage estimation procedure suffers from severe flaws, that render its results, and in particular statistical inference based on them, questionable. At the same time they propose a statistically grounded bootstrap based two-stage estimator that eliminates the above mentioned weaknesses of its na¨ıve predecessors and comes in two variants. This article introduces the new Stata command simarwilson that implements either variant of the suggested estimator in Stata. The command allows for various options, and extends the original procedure in some respects. For instance, it allows for analyzing both, output- and input-oriented efficiency. To demonstrate the capabilities of the new command simarwilson we use data from the Penn World Tables and the Global Competitiveness Report by the World Economic Forum to perform a cross-country empirical study about the importance of quality of governance of a country for its efficiency of output production., Discussion Paper / SFB823;11/2018

This paper analyzes whether moderate weight reduction improves subjective health perception in obese individuals. To cure possible endogeneity bias in the regression analysis, we use randomized monetary weight loss incentives as instrument for weight change. In contrast to related earlier work that also employed instrumental variables estimation, identification does not rely on long-term, between-individuals weight variation, but on short-term, within-individual weight variation. This allows for identifying short-term effects of moderate reductions in body weight on subjective health. In qualitative terms, our results are in line with previous findings pointing to weight loss in obese individuals resulting in improved subjective health. Yet, in contrast to these, we establish genuine short-term effects. This finding may encourage obese individuals in their weight loss attempts, since they are likely to be immediately rewarded for their efforts by subjective health improvements.

We complement the empirical evidence on the sustainability of weight loss achieved through cash rewards and, for the first time, rigorously examine the potential of cash rewards to prevent weight cycling. In a three period randomized controlled trial, about 700 obese persons were first assigned to two treatment groups, which were promised cash contingent on the achievement of an individually assigned target weight, and to a control group. Successful participants were subsequently allocated to two treatment groups offered cash rewards for confirming the previously achieved target weight and to a control group. This is the first experiment of this kind that finds effects of weight loss rewards up to 18 months after they were removed. Additional rewards only significantly improve the sustainability of weight loss while they are in place. Die empirische Analyse untersucht zum einen die Nachhaltigkeit einer Gewichtsreduktion, die mittels Geldanreizen erzielt wurde, und zum anderen die Wirksamkeit von Geldanreizen zur Verhinderung des Jo-Jo-Effekts nach einer erfolgreichen Gewichtsreduktion. In einem drei-phasigen Feldexperiment wurden ca. 700 fettleibige Teilnehmer zunächst zufällig auf zwei Anreizgruppen und einer Kontrollgruppe verteilt. Unabhängig von der Gruppenzugehörigkeit sollten ein individuelles Gewichtsreduktionziel erreicht werden, wobei nur Mitglieder der Anreizgruppen eine Geldprämie für das Erreichen des Zieles erhalten konnten. Erfolgreiche Teilnehmer wurden danach erneut zufällig einer Kontrollgruppe und zwei Anreizgruppen zugewiesen, wobei das Halten des Zielgewichts nur für letztere durch eine Geldprämie inzentiviert wurde. Anders als in früheren Experimenten, können Effekt der Abnehmprämie selbst 18 Monate nach ihrem Auslaufen nachgewiesen werden. Monetäre Anreize zur Verhinderung des Jo-Jo-Effekt wirken hingegen nicht über ihr Auslaufen hinaus.

The coexistence of social health insurance and private health insurance in Germany is subject to intense public debate. As only few have the opportunity to choose between the two systems, they are often regarded as privileged by the health insurance system. Applying a hazard model in discrete time, this paper examines the role of incentives set by the regulatory framework and the influence of individual personality characteristics on the decision to opt out of the statutory system. To address potential endogeneity of one of the key explanatory variables, an instrumental variable approach is also applied. The estimation results yield robust evidence on the choice of health insurance type that is consistent with pragmatic decision making, with both incentives set by regulation and personality traits as relevant determinants. Copyright © 2014 John WileySons, Ltd.

Obesity may not only be linked to undesirable health outcomes but also to limitations in sexual life. The present paper aims to assess whether there is a causal relationship between weight loss and sexual activity in adult obese individuals. To address the endogeneity of weight loss that is likely to result in biased estimation results, the analysis is based on data from a randomized field experiment. In this experiment financial weight-loss rewards were offered to a random subgroup of participants and can be used as exogenous source of weight variation in an instrumental variables approach. Estimation results indicate that for obese males loosing weight increases the probability for being involved in a sexual relationship. Conditional on having already lost some weight, a further reduction in obesity also increases the frequency of sexual intercourse. Übergewicht steht nicht nur im Zusammenhang mit Beeinträchtigungen im Gesundheitszustand, sondern ist auch mit Einschränkungen im Sexualleben assoziiert. Die vorliegende Arbeit analysiert, ob es einen kausalen Zusammenhang zwischen Gewichtsverlust und sexueller Aktivität bei erwachsenen übergewichtigen Personen gibt. Die Analyse basiert auf einem randomisierten Feldexperiment, um die Problematik der Endogenität des Gewichtsverlusts adäquat zu adressieren. In dem Experiment wurden finanzielle Anreize für eine zufällige Gruppe ausgelobt und als exogene Quelle von Gewichtsveränderung im Rahmen eines instrumentenvariablen Ansatzes verwendet. Die Schätzergebnisse zeigen, dass für übergewichtige Männer Abnehmen die Wahrscheinlichkeit erhöht, sich in einer sexuellen Beziehung zu befinden. Gegeben, dass bereits Gewicht abgenommen wurde, erhöht ein weiterer Rückgang des Übergewichts auch die Häufigkeit des Geschlechtsverkehrs.

This paper empirically assesses the relative role of health plan prices, service quality and optional benefits in the decision to choose a health plan. We link representative German SOEP panel data from 2007 to 2010 to (i) health plan service quality indicators, (ii) measures of voluntary benefit provision on top of federally mandated benefits, and (iii) health plan prices for almost all German health plans. Mixed logit models incorporate a total of 1,700 health plan choices with more than 50 choice sets for each individual. The findings suggest that, compared to prices, health plan service quality and supplemental benefits play a minor role in making a health plan choice. Dieses Papier untersucht empirisch die Rollen von Beitrag, Servicequalität und freiwilligen Zusatzleistungen im Rahmen der Krankenkassenwahl. Dazu verknüpfen wir repräsentative Paneldaten des SOEP der Jahre 2007 bis 2010 mit kassenspezifischen Indikatoren für (i) die Servicequalität, (ii) das Angebot von freiwilligen Zusatzleistungen und (iii) den Beitrag von fast allen deutschen Krankenkassen. Die verwendeten mixed logit Modelle basieren auf 1 700 individuellen Entscheidungssituationen, welche jeweils mehr als 50 Krankenkassen zur Wahl beinhalten. Die Ergebnisse deuten an, dass Servicequalität und freiwillige Zusatzleistungen eine - im Vergleich zu Beiträgen - untergeordnete Rolle in der Wahl der Krankenkasse spielen.

Nonrandom sample selection may render estimated treatment effects biased even if assignment of treatment is purely random. Lee (2009, Review of Economic Studies, 76: 1071–1102) proposes an estimator for treatment-effect bounds that limit the possible range of the treatment effect. In this approach, the lower and upper bound correspond to extreme assumptions about the missing information that are consistent with the observed data. In contrast to conventional parametric approaches to correcting for sample-selection bias, Lee’s bounds estimator rests on very few assumptions. I introduce the new command leebounds, which implements the estimator in Stata. The command allows for several options, such as tightening bounds by using covariates.

Non-random sample selection may render estimated treatment effects biased even if assignment of treatment is purely random. Lee (2009) proposes an estimator for treatment effect bounds that limit the possible range of the treatment effect. In this approach, the lower and upper bound, respectively, correspond to extreme assumptions about the missing information, which are consistent with the observed data. As opposed to conventional parametric approaches to correcting for sample selection bias, Lee's bounds estimator rests on very few assumptions. We introduce the new Stata command leebounds that implements the estimator in Stata. The command allows for several options, such as tightening bounds by the use of covariates., Discussion Paper / SFB 823;35/2013

The coexistence of social health insurance and private health insurance in Germany is subject to intense public debate. As only few have the opportunity to choose between the two systems, they are often regarded as privileged by the health insurance system. Applying a hazard model in discrete time, this paper examines the role of incentives set by the regulatory framework as well as the influence of individual personality characteristics on the decision to opt out of the statutory system. To address potential endogeneity of one of the key explanatory variables an instrumental variable approach is also applied. The estimation results yield robust evidence on the choice of health insurance type that is consistent with rational decision making, with both incentives set by regulation and personality traits as relevant determinants. Die Koexistenz von gesetzlicher und privater Krankenversicherung in Deutschland ist Gegenstand intensiver öffentlicher Debatten. Da lediglich eine Minderheit der Versicherten die Wahl zwischen beiden Systemen hat, wird diese Minderheit oft als privilegiert betrachtet. Auf Basis des Sozioökonomischen Panels (SOEP) und unter Verwendung eines Hazard-Models in diskreter Zeit untersucht dieses Papier die Wirkung von Anreizen, die sich aus den gesetzlichen Rahmenbedingungen ergeben sowie den Einfluss von individuellen Persönlichkeitsmerkmalen auf das Wechselverhalten. Um die potenzielle Endogenität des Gesundheitszustandes zu berücksichtigen, wird außerdem ein Instrumenten-Variablen Ansatz verwendet. Die empirischen Ergebnisse liefern robuste Evidenz, dass die Wahl des Krankenversicherungssystems konsistent mit rationalem Entscheidungsverhalten ist und sowohl durch gesetzliche Anreize als auch durch individuelle Persönlichkeitsmerkmale determiniert ist.

The question of whether alcohol and tobacco are consumed as complements or substitutes is crucial for determining the side-effects of anti-smoking policies. Numerous papers have empirically addressed this issue by estimating demand systems for alcohol and tobacco, and subsequently calculating cross-price effects. However, this traditional approach is often seriously hampered by insufficient price variation observed in survey data. We, therefore, suggest an alternative instrumental variables approach that statistically mimics an experimental study and does not rely on prices as explanatory variables. This approach is applied by means of German survey data. Our estimation results suggest that a reduction in tobacco consumption results in a moderate reduction in alcohol consumption. It is demonstrated that this implies that alcohol and tobacco are complements. Hence, we conclude that successful anti-smoking policies will not result in the unintended side-effect of an increased (ab)use of alcohol.

The substitutability of inputs to health production in hospitals has recently become an issue in Germany. New regulations concerning the renumeration of hospitals set incentives to reduce patients’ length of stay and, in turn, to operate less care-intensive. Based on data at the individual hospital level from Germany, covering the years 2006 to 2008, we test whether substitutability of inputs and its change over time exhibits heterogeneity across different types of ownership. If found in the data, this may point at differently owned hospitals adapting differently to the new regulations. In order to avoid relying on input prices as regressors, which exhibit only very limited variation across hospitals, we pursue a price-free empirical approach by calculating technical elasticities of substitution from the multiplier representation of data envelopment analysis. The empirical analysis yields pronounces ownership-specific heterogeneity in the technical elasticity of substitution between physicians and nurses indicating that non-profit hospitals operate particularly ‘physician-intensive’. Yet, this pattern is stable over the considered period., Discussion Paper / SFB 823;41/2012

The classical Heckman (1976, 1979) selection correction estimator (heckit) is misspecified and inconsistent if an interaction of the outcome variable and an explanatory variable matters for selection. To address this specification problem, a full information maximum likelihood estimator and a simple two-step estimator are developed. Monte-Carlo simulations illustrate that the bias of the ordinary heckit estimator is removed by these generalized estimation procedures. Along with OLS and the ordinary heckit procedure, we apply these estimators to data from a randomized trial that evaluates the effectiveness of financial incentives for weight loss among the obese. Estimation results indicate that the choice of the estimation procedure clearly matters., Discussion Paper / SFB 823;40/2012