In France as in most developed countries, designing an efficient system of public financing for long-term care is one of the most striking challenges of the next two decades. But the deep lack of knowledge we face make the task rather uncertain. In this context, the main aim of this project is to establish some quantitative empirical evidences about how economic parameters, such as the private cost of formal home help for disabled elderly, impact their care arrangements. Getting an estimation of the price-elasticity of the demand for formal home help and of its effect on informal care will allow us to simulate the consequences of some reforms of the public financing scheme on care arrangements (quantity of home help provided by formal and informal caregivers), on the amount of public resources allocated to the elderly and on its distribution among them, according to income and disability level. Method The scarcity of econometrical research devoted to the impact of the cost of care in ageing economics is mainly due to the difficulty to collect precise data on prices and out of pocket expenditures, since, in many countries, this field of public policy is decentralized: the way public benefits are calculated thus differs according to the geographical location of the claimant. The French minister for health and social affairs will produce next year a specific survey, called CARE, which will include all the information needed to estimate our parameters of interest with standard econometric strategies, but will not be renewed afterwards. We thus propose, first, to use this survey to estimate the impact of private cost of formal help on care arrangement with standard models and methods and, second, to use the results as a benchmark for original alternative strategies, which require less specific data but rely on more sophisticated methods (partial identification) and take advantage on a precise knowledge of the decentralized rules of public financing. These alternative strategies will be implemented with administrative surveys used routinely for the follow-up of APA beneficiaries, the Handicap-Santé survey, which is renew every ten years, and a specific survey, called "Territories", conducted by our research team in 2012, in which we collected information from 73 metropolitan Conseils généraux on their practices regarding home services price-setting and demand subsidization. This gives to the project a methodological dimension since we will be able to evaluate estimation strategies, which could be usefull for analysing other fields of decentralized public policies. Working program Three alternative strategies will be implemented in the first part of the project : 1/ Estimating the price-elasticity of the demand for formal home help, using data of one specific department: in this case, the Territoires survey allows us to calculate the exact private cost of one hour of home help for APA beneficiaries cared by a service whose price was set by the local government. We already have the result for one département and will work on another one to test the robustness of our results. 2/ Estimating the price-elasticity of the demand for formal home help, using a sample of APA beneficiary gathered by the Health minister across 66 départements and partial identification methods : in this case, information is too poor to calculate the exact individual private cost, but the Territoires survey will allow us to establish a range of cost for each department. 3/ Estimating the impact of the amount of formal home help on informal care, through a bivariate model, using the Handicap Santé Survey. The Territoires survey will in this case offer proper instruments for the amount of formal home help in order to deal with the simultaneity of decisions regarding formal and informal care. In the second part of the project, we will use our estimation results to build a simulation model and analyze the impact of various reforms of price-setting and subsidization.