1. Card et al. (2008) uses a natural experiment that uses eligibility criteria (age of 65) as an identification
strategy to identify the effects of Medicare on mortality outcomes.
a. Explain in detail how the eligibility criteria set for Medicare is used as a quasi-natural experiment.
Talk about the approach (method) they use.
b. The analysis of Card et al.’s study is based on hospital level data such that people in the sample
are those who are admitted to the hospital. This creates a selection problem if we compare people
who are right above the threshold vs. those who are right below the threshold. This is because
if a 64 year old plays a waiting game and goes to the doctor when she is 65, she would have
waited longer compared to a 64 year old with similar illness but who does not play the waiting
game (i.e., goes to the doctor when 64). Hence, comparing the health outcomes of 64 year old
individuals (without Medicare) vs. 65 year old individuls (with Medicare) might mean that we are
comparing two groups that are systematically different in health stock to begin with. Describe
how the researchers get around this selection problem. (hint: hospital admissions are higher in
week days compared to weekends.)
c. By using a figure that corresponds to the eligibility criteria (age), briefly describe the findings of
their study on: i) insurance coverage, ii) quality of care, and iii) mortality outcomes.
d. What are some potential drawbacks of their study?