I’ve got some original-ish research to report on.
Among the many controversial aspects of expanding medicaid to more Americans, there are debates on whether medicaid:
- Is low quality care that ends up making its recipients worse off
- Causes a decrease in the labor supply since some workers on the margin will be less inclined to find work if they already have health insurance
- Causes significant increases in certain healthcare utilization like emergency room visits that cause a strain on the system
- Causes an increase in happiness merely by reducing the financial stress of being uninsured
It can be tough to totally figure out the magnitude of these effects since comparing a population that has medicaid and a population that doesn’t can’t account for unobserved characteristics that will almost certainly bias estimates. But luckily, we have a social scientist’s dream.
Oregon had decreased its medicaid rolls due to budget cuts and then in 2008 realized they had funding for 10,000 extra spots. Knowing they’d have more interest than availability, they left these spots up to a random lottery that ended up having ~90,000 applicants. Since the chosen households would be picked at random, it’d be possible to compare the population of those who were selected by the lottery to those who didn’t and evaluate their outcomes.
The NBER has already done a lot of good research on this. Their initial findings suggest that after being on Medicaid in Oregon’s Oregon Health Plan for 12 months increases medical utilization (for better or worse), decreases financial strain, and increases self-reported measures of happiness. One significant finding: “…if we compare our estimates to the literature on the impact of income on happiness, the impact of insurance is roughly equivalent to the impact of a doubling of income.”
But what about the effect on labor supply? The CBO and Casey Mulligan have suggested Obamacare will significantly reduce the number of workers in the labor force. Obamacare covers the entire country and the OHP only covers a subset of the Oregon population, so this is not an entirely apples to apples comparison. Nonetheless, I can say from doing initial estimates using the data from the OHP study that, controlling for a variety of individual and household characteristics, being on OHP because of the lottery has no significant effect on employment compared to those who were in the lottery and were not selected.
There has been some work getting to this conclusion, but I haven’t found any that goes as in depth as I have looked. For example, one theory is that this expansion could cause older people to exit the labor force since the only thing keeping them in a job until they’re medicare-eligible is the presence of healthcare. I have so far found that separating on quantiles of age does not produce significantly different results. The same can be said of dividing up the data into quantiles based on household income – the idea that people approaching the income threshold for medicaid will stop working to continue staying on government assistance has also been so far refuted.
The findings still need to be reviewed, but I think it’s a notable find.