Thread by @smilleralert

It's well known that lower income people tend to suffer much worse health. Is poverty at the heart of this disparity, and, if so, could a large cash transfer help close this gap? We examine an RCT that provided 1000 low income participants $1000/month for 3 years. We find…

The cash generated big improvements in stress and mental health, but they were short-lived. By the second year of the transfer, treatment and control reported similar rates of stress and mental health, and we can rule out even small improvements.

There was no effect of the transfer on physical health, measured via self-reports, clinical outcomes derived from blood draws, and admin records of mortality. For the former two, we can rule out even very small improvements (the mortality effects are more noisy).

Surprisingly little effect on self-reported access to medical care. No effect on exercise or sleep. Again our confidence intervals rule out even very small effects.

Mirroring what we see for mental health, the transfer generated large but short-lived reductions in food hardship/food insecurity; even by year 2, no significant difference across treatment arms.

But! The cash did lead to people using more medical care: hospitalizations and ER visits go up, dental care did as well, and spending on medical care goes up by about $20/month. Possible that this higher usage could have improved health over a longer time horizon.

Suggestive results: frequency of drinking alcohol goes up (perhaps due to more socializing), but reports of drinking interfering with responsibilities goes down, as does reported abuse of painkillers. (Note these results don't survive an adjustment for multiple testing)

There's so much energy in health policy now for addressing "social determinants of health"--and poverty in particular. Could cash transfers be the way to meaningfully and effectively reduce health disparities? It's hard for me to look at these results and say yes.

What's great about cash is it gives the freedom to choose what you want to consume! But that also makes it a rather blunt tool for addressing health specifically. Our participants consumed more leisure, food, housing, and other stuff. And different people chose different things.

Those consumption choices did not appear to improve their health on average, but they were the things participants wanted, as revealed by their own choices. This is a feature of cash, not a bug!

If the goal policy is to improve health specifically, there are health-targeted interventions that we know work--make medical care cheaper, expand coverage, reduce barriers to initiating a primary care relationship.

But if your goal is to reduce poverty or even just to give people the freedom to consume what they want, cash transfers are still an important tool in the policy portfolio.

Paper:

Joint work with @elizabethrds @AlexBartik @dbroockman @evavivaltpublic.websites.umich.edu/~mille/ORUS_He…