Geographical equity in decentralized policies: theory and application to care for dependent elderly people – EQUIDEC-2

How can we assess the geographical equity of public policies affecting essential living conditions (health, social, education, etc.) when they are partly decentralised to local authorities? This is the central question of the EQUIDEC-2 research project, which combines two dimensions. A theoretical dimension aims to develop criteria for assessing geographical equity that are compatible with decentralisation, for different theories of justice. The type of policy we are interested in leads us to adopt a third way between the two main options structuring the literature: setting constraints on the absolute level of services provided (and not relative to effort as in the Buchanan approach), while respecting local specificities in terms of preferences (contrary to operational planning practices). In a second dimension, we will apply the approach to the specific case of French policies for the so-called "dependent" elderly, which are largely decentralised to departmental councils. This operational dimension involves the use of a large number of existing databases and the production of two original surveys.
Previous research (EQUIDEC) has identified four types of theories of justice adapted to our question within a framework proposed by Fleurbaey: they propose to distribute either a specific distribuendum (specific resources) or a complete distribuendum (well-being or capabilities) and are based on either egalitarian or sufficientarist criteria. This research has also allowed us to test the empirical feasibility of calculating an indicator of geographical and financial accessibility to nursing homes (Ehpad). Last but not least, it has raised new questions, some of which motivate this new project.
Three seminars, involving academics and policy-makers, will monitor the progress of the project (WP1).
We propose three methodological innovations for the definition of indicators measuring accessibility to care (WP2). The first deals with the reduced access experienced by certain categories of the population according to their relative characteristics: the aim is to integrate selection for admission to nursing homes (Ehpad) in a context of drastic control of the supply of places. The other two concern the calculation of accessibility rates aggregating several specific distribuenda. To take account of the need to have several types of care (nursing care and home help) available simultaneously in order to be able to remain at home in decent conditions (complementarity), we propose a method based on care baskets. To address the fact that institutional care and home care are not used at the same time (they are mutually exclusive), we propose a method that calculates competing populations based on simulated demand behaviour.
Two surveys will be conducted to measure the distributions of well-being and capabilities of elderly people with limitations in activities of daily living as a function of their individual and environmental characteristics and the type of care they receive (WP3). The first survey will provide ICECAP-O valuation scores, which are not yet available for France, as well as a full autonomy equivalent income index. We will then be able to use the data from a second survey to measure the well-being and capabilities of a representative sample of the frail population in 4 départements, to estimate conditional distributions and use them to simulate the levels of well-being and capabilities in the other départements.
Finally, the indicators developed in WP2&3 will be used to assess the geographical equity of care provision (WP4). The assessment will be carried out by applying egalitarian or sufficientarist criteria to three types of distributions: accessibility rates; well-being or capabilities, according to the actual mode of care; opportunities for well-being or capabilities. Finally, to broaden the scope of the evaluation, we will explore the construction of social welfare functions to include informal caregivers and taxpayers.