In times of austerity we should think hard about welfare policy. Or even better, we should study the data. Most of it comes from the USA, perhaps because in the UK we don't think it's researchable, or perhaps because 50 states with different policies, and policies that keep changing, generate plenty of variation to study.
So here are two interesting papers.
The first one looks at the effect of more generous unemployment benefit. The analysis is a regression model and the outcome of interest is self-rated health, collapsed into a single number: the percent of people rating their health as 'fair' or 'poor'. That percentage is higher for unemployed than employed, of course, but the gap narrows as unemployment benefit increases. Interestingly that's only true for men. It's an ecological study so all we can say is that the gap narrows in states (and at times periods) with higher levels of benefit.
The second study addresses help for families with children. A series of randomised trials in the USA compared indefinite welfare benefits (Aid to Families with Dependent Children AFDC) with time limited benefits.
"These randomized controlled trials found not only that time limits to cash assistance incentivized participants to move into the workforce, but also that they produced increases in earnings relative to traditional AFDC.Ultimately, these experiments contributed to the passage of the Personal Responsibility and Work Opportunity Reconciliation Act in 1996. This act was perhaps one of the most sweeping US policies enacted within the past 2 decades and one of the few large-scale policies to be passed on the basis of a large and convincing body of scientific evidence."
I must admit (to my shame) that I was unaware of this research till now, but it does explain a lot of right-wing rhetoric. Time limited benefits may motivate people who are capable of work to get paid jobs; but they also disadvantage people who simply can't enter the workforce. It's this problem which the second study explores, by means of a simple economic model. The model is populated with data from two trials. The result is that indefinite benefits cost a society more but result in lower infant mortality; so you can calculate a cost per life saved: an incremental cost effectiveness ratio.