Informing policy priorities using inferencefrom life satisfaction responses in a large community survey

Compensating differentials were thus calculated for each of the variables in the model as follows:1 1To formalize the idea of compensating differentials for changes in a discrete, or indicator, variable Xi, note that the coefficient bi in (1) represents an estimate of the discrete change ∆LS resulting from a unit change in Xi. [...] The estimated benefits from improving health or one of the other life experience variables in the model either exceed, or are of comparable size, to the effect of traversing the entire income range. [...] When improvements are made to life experience variables, the aggregate increase in life satisfaction at the population level depends on both the strength of the effect for a given individual (table 3) as well as the number of individuals within the population that would be affected by the change. [...] Similarly, the perceived responsiveness of local-government to the needs of residents emerged as a strong contrib- utor to improved life satisfaction with a predicted increase in life satisfaction of 7.7 out of 100 for the average survey respondent who moves from feeling that local-government is “not at all” responsive to the needs of residents to feeling that the government’s re- sponsiveness is. [...] Our results show that in addition to the quality of the government, the per- ceived responsiveness of government to the needs of residents emerged as the single most important experience variable explaining differences in life satisfaction (excluding physical and mental health).