Close
Ian Walker and Maria Navarro Paniagua, University of Lancaster and David Stott, University of Glasgow
(Project no. 0301565, previously 30156)
In this project we will be looking directly at the reduced form relationship between Winter Fuel Payments (WFP) and mortality - using the England and Wales Longitudinal Study merged with local level Met Office data. The reduced form will consists of exploring mortality as a function of age, meteorological data, individual's characteristics available in the LS data and for the evaluation of the policy a variable that will indicate the amount of WFP the household the individual lives in is entitled to. First, we will model the survival hazard (how mortality varies with age) and the probably highly nonlinear impact of cold. Second, we intend to implement Regression Discontinuity Design and Differences in Differences Analysis to evaluate the policy.
The work will exploit the regression discontinuities associated with the age related element of WFP, and natural experiments associated with reforms to the levels of WFP that have occurred over time, differentially for different age ranges. In the course of the work we will also be modeling the survival hazard (how mortality varies with age) and hence the (probably highly nonlinear) impact of cold. Although the LS datasets are collected at four (England and Wales) census dates, mortality information (date, cause and place of death) from the Vital Statistics data has been merged into the LS datasets on a continuous basis. Usefully the introduction of WFP occurred in 1997 and was greatly expanded in 2000.
This is an opportune time to investigate WFP because there is no evidence on its effectiveness in achieving its ostensible objective and it constitutes a large item of government spending at a time when there is considerable pressure on public spending. Moreover, new policy initiatives are starting and this research will enable relative cost effectiveness judgements to be made. We know exactly the amount of WFP a household receives since this depends on the number of cohabitants being above a certain age threshold (60 years old). The amount increases by age and has varied since it was first introduced. The take up rate is almost 100%.