Where Does Your Health Care Dollar Go?


May 22, 2018

 Your premium—how much you pay for your health insurance coverage each month—helps cover the costs of the medications and care you receive. It also helps to improve health care quality and affordability for all Americans. Here is where your health care dollar really goes.

Download the full infographic.



The objective of this analysis is to understand how premiums for a typical commercial health insurance plan are invested.

Specifically, the analysis shows the inflation-adjusted average annual amounts paid by commercial health insurance plans in 2014-2016 for the medical care of plan members; the average annual amounts paid for general operating expenses of the health plan in 2014-16; and the average annual reported profit (or loss) in 2014-16.

To determine the annual amounts paid for medical services in 2014-16, the commercial claims data from the MarketScan® Commercial Claims and Encounters database (Truven Health Analytics, Inc) were summarized. The Inpatient Services file, the Outpatient Services file, and the Outpatient Drug file of the MarketScan® database were utilized for the study.

AHIP randomly selected a total sample of 30 health plans: five for-profit and 25 not-for-profit plans.  Plans were selected from a list of all health plans that had the majority of their business (i.e., greater than 50% of enrollees) in the commercial market, based on enrollment data from AIS’s Directory of Health Plans: 2014 (AIS). The not-for-profit plans sample was further reduced by excluding three plans that had just started operations in 2014, one plan that ceased operations as an independent entity during the study period, and four plans that lacked publicly available statutory information.

To estimate operating costs and profitability, for the five publicly-traded insurance companies, their 2014-2016 10K filings with the Securities and Exchange Commission were examined; while operating expense and profitability data for private, not-for-profit organizations were extracted from their 2014-2016 Form 990s, filed with the Internal Revenue Service. We were unable to obtain a Form 990 for one plan in 2014 and two plans in 2016, thus, their operating expenses and profits are the average of two-years of data.

Since the analysis used multiple years of data, all expenditure data from 2014 and 2015 were adjusted for inflation and expressed in 2016 dollars. This inflation adjustment was performed using the Medical Care Component of the Consumer Price Index (CPI) reported by the U.S. Bureau of Labor Statistics (www.bls.gov).

Only those patients under the age of 65 on the date of service who had evidence of continuous health plan enrollment for the entire period in each study year (2014, 2015, or 2016) and had prescription drug coverage were included. Claims having missing payment information; missing dates of service; and in the case of the inpatient and outpatient services claims, missing data on whether the claim was submitted by the facility or the physician were excluded from the study. The main variable of interest was the “net payment” variable which is the amount paid by the health insurer. The net payment amounts of all included claims for each study year were summed

Only those revenues attributable to premium payments from health plan members were recorded for each plan for each year (2014-2016). Revenues from sources other than premium payments, such as from other business segments or investment income, for example, were excluded.  For each plan, the average revenue across the three years was calculated.

For the five publicly-traded, for-profit insurers, amounts listed in their 10K filings as “Operating Costs,” “General and Administrative Expenses” or “Sales, General and Administrative” were extracted from their 2014-2016 Income Statements.  Amounts paid for taxes were also recorded.  Amounts shown as “Net Income” or “Net Profit” were also recorded. For each plan, an average total operating cost and an average net profit were calculated across the three years and recorded.

For the seventeen private, not-for-profit entities their total operating expenses were calculated by subtracting the “Benefits Paid to or For Members” from the “Total Functional Expenses” amounts appearing in their Form 990. Similarly, profitability was determined by subtracting the “Total Expenses” from their “Total Revenues”. These calculations were performed for each plan for each year and recorded. For each plan, an average total operating cost and an average net profit were calculated across the three years and recorded.

Finally, for those health plans having multiple revenue streams beyond member premiums (such as provision of business-to business services), some of the plans’ total operating expenses and profits could be unrelated to servicing their insured population. To account for that, we apportioned the operating expenses and profits based on the share of health plan’s revenue derived from member premiums. For example, if a plan A had 80% of its revenue derived from member premiums, we used 80% of its total operating expenses and profits in our calculations.

For each plan, the average total operating expense and the average net profit amounts were divided by the average revenues derived from premiums to yield that insurer’s operating margin and net profit margin. To account for differences in the sampling of for-profit (n=5 plans) and not-for-profits (n=17 plans), a simple average of the operating margin and an average of the net profit margins were calculated separately for the for-profit and not-for-profit plan subgroups. These two averages were then weighted by the share of commercial enrollment for the publicly-traded, for-profit plans and the not-for-profit plans in 2014-16 based on AIS data and combined.

The average total operating expense calculated across all plans was further subcategorized into the key functional areas summarized in Table 2.  The proportions of the average total operating costs attributable to each of these core administrative functions was determined by the consulting firm, Milliman, Inc., using their proprietary administrative effectiveness and efficiency benchmarking model. This proprietary benchmarking model is based on data collected from more than 100 health plans and third-party administrators. For this study, the model was adjusted to reflect the study-specific health plans sampled with respect to plan size, enrollment, and business mix.