Insurers , , , and randomly select 308,000,000; 10,000,000; 1,000,000; 100,000; and 10,000 policyholders from the same population (See Table 1 Row 1). Each insurer produces a Population Loss Ratio Estimate (PLRE) of the Population Loss Ratio. How accurate these PLREs are is measured by each insurer's portfolio size adjusted standard error. Insurers select large samples, so their PLRE distribution functions are all normally distributed, even if policyholder PLREs are not. I will show that transferring insurance risks to health care providers (Capitation) is flawed, by comparing the impact of portfolio size on PLRE variability, PLRE probabilities, and these five insurers' probabilities of earning Profits, incurring Operating Losses, or becoming Insolvent. I will also compare insurers' Surplus requirements and Maximum Sustainable Benefits for policyholders.