A recent paper by Hai Zhong (2011) finds that health insurance that provides immediate reimbursement for health care services significantly increases the likelihood of patients seeking outpatient treatment in China compared to reimbursement beneficiaries with a delay. China isn’t the only country where insurance companies provide delayed reimbursement. In fact, in France patients pay the full cost of physician visits up front and only later are reimbursed 70 percent of the cost.
Why would the delayed reimbursement make a difference? I can think of three reasons.
- Liquidity Constraints. Some individuals may not be able to afford the payment. Poor individuals may literally not have the capital to pay for these services up front. Getting loans from formal institutions (e.g., banks) or informal ones (e.g., friends and family) may be costly either in terms of interest of obligations to family and friends. Even if an individual is rich, acquiring extra money may be costly (e.g., trip to ATM, ATM fees, interest on credit card).
- Probability of Non-Payment. Although may policies are written where payment is assured, in practice reimbursement rates will not be 100 percent. For instance, individuals could fail to submit the correct forms for reimbursement, they could move addresses, or the patient could die. In addition, patients may have some uncertainty surrounding the benefits covered and thus they may not be 100% sure that they will receive reimbursement. Beneficiaries may not trust their insurance plan; they may assume it is trying to cheat them and thus with some non-zero probability the beneficiary will not get paid.
- Reflection of value. Even if a patient is rich and payment probabilities are 100%, the patient may still be less likely to use the service if they don’t need it if they realize the true cost. Alternatively, patients who realize a service is valuable may also be more likely to use it.
- Zhong, H. (2011), Effect of patient reimbursement method on health-care utilization: evidence from China. Health Economics, 20: 1312–1329. doi: 10.1002/hec.1670