On the fee-for-service model

It increases access to health care, but drives up costs:

This is the only form of payment under which the provider does not have any incentive to select healthier patients; in fact, the opposite is true.  Under FFS, the provider bears no risk for the cost of treatment.  The payer, the insurer, or the patient is entirely at risk for the cost of care.  Theoretically, patients and third-party payers have a reason to question the need for additional services and negotiate lower payments.  In reality, however, patients and third-party payers can seldom negotiate effectively due to the professional power of physicians.  

Studies in many countries, both developed and developing, have found that a fee-for-service system promotes an excessive use of services and an increase in costs (Barnum et al. 1995).  Comparing resource utilization under two provider-payment methods (FFS and capitation) in Thailand, Yip et all (2001) found a significant difference in the average length of stay, drug charges per admission, and lab costs per case.  Under FFS, resource utilization was consistently greater.  A study in the United States (Krawelski et al. 2000) found similar results.  Costs were significantly lower under medical group practices paid by capitation than those paid by FFS.  

That's from Roberts and Hsiao, in their chapter on Payment.