Zero‐inflated Endogenous Count in Censored Model: Effects of Informal Family Care on Formal Health Care

Abstract

If informal family health care is a substitute for formal health care, then there is a scope to reduce formal health care cost by promoting informal family health care. With the use of Korean data for the elderly, this paper estimates the effects of informal family health care on formal health care, where the former is measured by the number of caregivers and the latter is measured by the formal health care expenditure. This task, however, poses a number of difficulties. The first is that the number of the family caregivers is an endogenous count regressor. The second is that there are too many zeros in the count (85%). The third is that the response variable also has a nontrivial proportion of zeros (14%). This paper overcomes these problems by combining 'control function approach', 'zero-inflated' counts, and a semiparametric estimator for censored models. The resulting procedure avoids strong parametric assumptions and behaves well computationally. Our main empirical finding is that informal family health care has a large substitute effect for diabetics and that there are also weak evidences that informal family health care has substitute effects for high blood pressure and mental diseases.

Treatment Effect Estimation Using Nonlinear Two‐Stage Instrumental Variable Estimators: Another Cautionary Note

“…However, recent simulation research by Terza and colleagues has suggested that IV‐based nonlinear two‐stage residual inclusion (2SRI) estimators—a special case of control function methods (Hausman ; Garrido et al. ; Lee and Kim )—can directly estimate the population average treatment effect (ATE) in models with binary, or otherwise “inherently nonlinear,” dependent variables without relying on these assumptions (Terza, Bradford, and Dismuke ; Terza, Basu, and Rathouz ). As a result, the authors of this research suggest that nonlinear 2SRI should always be employed for analysis of empirical models with inherently nonlinear dependent variables and unmeasured confounding risk (Terza, Basu, and Rathouz ).…”

Estimation of Binary Response Models With Endogenous Regressors

“…Almirall et al . ; Lee and Kim, ); however, this requires some extra modelling effort. Because terms such as v 2 or xv 2 are explicitly accounted for, the error term variance tends to get reduced in the CF approach, which comes at the cost of correctly modelling the function λ ( v 2 ; θ ) or λ ( v 2 , x ; θ ) .…”

Investigating the relationship between formal and informal care: An application using panel data for people living together

“…The substitution effect weakened as the level of recipient disability increased. Lee and Kim () found substitution effects that varied in magnitude by the morbidity of the recipient. Gannon and Davin () used SHARE data for France and Ireland and examined a substitution relationship between informal care and formal home care for those over 50 years old.…”