Private, Social, and Self‐Insurance for Long‐Term Care in the Presence of Family Help

• Author: PH. DE DONDER,P. PESTIEAU
• Published date: 01 February 2017
• Date: 01 February 2017
• DOI: http://doi.org/10.1111/jpet.12163

PRIVATE,SOCIAL,AND SELF-INSURANCE FOR LONG-TERM CARE

IN THE PRESENCE OF FAMILY HELP

PH. DE DONDER

Toulouse School of Economics

P. PESTIEAU

HEC-Management School University of Li`

ege, Universit´

e catholique de Louvain, TSE, and CEPR

Abstract

We study the political determination of the level of social long-term

care insurance when voters can top up with private insurance, saving

and family help. Agents differ in income, probability of becoming de-

pendent and of receiving family help, and amount of family help re-

ceived. Social insurance redistributes across income and risk levels,

while private insurance is actuarially fair. The income-to-dependency

probability ratio of agents determines whether they prefer social or pri-

vate insurance. Family support crowds out the demand for both social

and, especially, private insurance, as strong prospects of family help

drive the demand for private insurance to zero. The availability of pri-

vate insurance decreases the demand for social insurance but need not

decrease its majority-chosen level. A majority of voters would oppose

banning private insurance.

1. Introduction

While health care services aim at changing a health condition (from unwell to well),

long-term care (hereafter LTC) merely aims at making the current condition (unwell)

more bearable. Individuals need LTC due to disability, chronic condition, trauma, or

illness, which limit their ability to carry out basic self-care or personal tasks that must be

performed every day. Such activities are defined as activities of daily living (eating, dress-

ing, bathing, getting in and out of bed, toileting, and continence) or instrumental activ-

ities of daily living (preparing own meals, cleaning, laundry, taking medication, getting

to places beyond walking distance, shopping, managing money affairs, and using the

telephone/Internet). A person is dependent if he or she has limitations in either type.

Ph. De Donder, Toulouse School of Economics (CNRS-GREMAQ and IDEI), Manufacture des

Tabacs, MS102, 21 all´

ee de Brienne, 31015 Toulouse Cedex 6, France (philippe.dedonder@tse-fr.eu).

P. Pestieau, CREPP, HEC-Management School University of Li`

ege; CORE, Universit´

e catholique de

Louvain; TSE and CEPR, Boulevard du Rectorat, 7 (B 31) 4000 Li`

ege, Belgium. (p.pestieau@ulg.ac.be)

A former version of this paper has been presented under the title “Voting over LTCPublic Insurance”

in Munich (CESifo area conference on public sector economics), Toulouse, and Nashville (Vanderbilt

University conference on tax theory). We thank participants as well as Helmuth Cremer, Katherine

Cuff, and Erik Schokkaert and three referees for their comments. Financial support from the Chaire

“March´

e des risques et cr´

eation de valeur” of the FdR/SCOR is gratefully acknowledged.

Received March 3, 2015; Accepted March 20, 2015.

C2016 Wiley Periodicals, Inc.

Journal of Public Economic Theory, 19 (1), 2017, pp. 18–37.

18

Long-Term Care Insurance 19

Dependent people can draw on four different types of resources to help alleviate

their daily living problems. By far the most important quantitatively are their own re-

sources (self-insurance,1or savings) and family help (mainly through informal help).

Several countries also offer some form of social LTC insurance, although the size of

these programs is usually low, especially compared to the pension programs. Finally,

except for a handful of countries (such as the United States and France), private insur-

ance’s role is negligible, and in any case consistently smaller than that of the State.

Several articles have studied the interactions between some of these resources (see

Section 1.1). The value added of our paper is that it considers simultaneously the four

types of resources (family help, private, social, and self-insurance) and that it studies the

political choice of the level of social LTC insurance. More precisely, we study the deter-

minants of the political support for social insurance in an environment where people

differ in income, risk, and availability of family help, and where they choose individually

their private and self-insurance levels. As stated above, the availability of family help is

of first importance for LTC, and distinguishes our approach from the literature study-

ing the political support for other kinds of social insurance programs, such as health or

social security.2

We develop a framework where agents live two periods. They earn an income, pay

taxes, save and buy private LTC insurance when young. Beyond income, they also differ

in the probability of becoming dependent when old, in the probability of receiving

help from their family if dependent, and in the amount received from the family.3They

choose by majority voting the value of the proportional income tax rate that finances

the lump sum social insurance transfer received if dependent. A crucial assumption is

that social LTC insurance redistributes across income and (ex ante) across risk, while

private insurance is actuarially fair and not redistributive.4To level the playing field,

both forms of insurance are equally efficient (no loading factor, no cost of public fund,

or distortionary impact from taxation).5

While we endogenize the levels of the three types of insurance, family help is mod-

eled as an exogenous norm. This corresponds to what Costa-Font (2010) calls familism,

a short hand for “familism culture,” or “the embeddedness in a family’s social norms

(family ties)” (p. 2). Familism is exogenous because of the facts that “need of LTC is

a non-repeated experience in human life, and that individuals are arguably prone to

1Like other papers dealing with LTC (see Costa-Font 2010, for instance), we use the general definition

of self-insurance (assets set aside to cover possible losses) rather than the more specific meaning used

in insurance theory (prevention activities reducing the severity of the potential loss).

2For instance, over 80% of dependent elderly live in their home or with their children, and for these

people most of the care is informal. See Stone (2000).

3Like most related papers, we assume that the state of the world for each individual is binary, with

one being either dependent or not. In reality, there are several degrees of severity in the state of de-

pendence. Public and private insurance schemes use scales such as the Katz or IADL (Independence

in Activities of Daily Living) index to measure those degrees and provide compensations accordingly.

Introducing such scales in our analysis would complicate matters a lot without bringing significant new

insights.

4From now on, we slightly abuse terminology and refer to the risk of an agent as his probability of

becoming dependent.

5Alternatively, we could have modeled public intervention as the subsidization of private insurance

premia. Such a policy also benefits high-risk agents, provided that the demand for private insurance is

increasing in individual risk. Its income redistribution component is less clear: richer agents pay more

taxes, but may also consume more insurance and thus benefit more from the subsidy. We leave this

analysis for future research.