Estimating Federal Payments and Eligibility for Basic Health Programs: An Illustrative Example
Stan Dorn and Jennifer Tolbert. The ACA’s Basic Health Program Option: Federal Requirements and State Trade-Offs, November 2014, Washington, DC: Kaiser Family Foundation and Urban Institute.
How Federal BHP Payment Amounts are Determined
CMS. “Basic Health Program: Federal Funding Methodology for Program Year 2016,” Federal Register, October 23, 2014, Vol. 79, No. 205, pp. 63363- 63376, http://www.gpo.gov/fdsys/pkg/FR-2014-10-23/pdf/2014-25257.pdf.
After a BHP program’s first year, the federal government will need to make an additional adjustment to the reference premium, captured by the Population Health Factor (PHF). At that point, BHP enrollees will no longer be in the individual market. They may have a different average risk level than the remaining participant's in the individual market. If so, premiums charged in the individual market without the participation of consumers under 200% FPL might be different than if BHP enrollees had stayed in the individual market. The PHF will adjust marketplace premiums to compensate for the change in risk levels made by the removal of BHP consumers so the reference premium reflects what would have been charged without BHP. If BHP consumers are healthier, on average, than individual market participants, the PHF will reduce the premium from levels observed in the marketplace. If they are less healthy, it will raise the premium.
In future years, it should not be difficult to determine the PHF. Each individual market participant's risk level will be measured as part of the risk adjustment system. If states gather similar information about BHP enrollees, actuaries should be able to estimate the impact on individual market premiums if BHP-eligible consumers were added to the individual market.
In most states today, the PHF is even easier to calculate for the first year of BHP program operation: it does not affect premiums at all. That is because, in 2014, BHP-eligible consumers are already in the individual market in most states. As a result, marketplace premiums are already based on the risk pool that would apply without the operation of BHP.
However, in 2014 a handful of states—especially Minnesota, which covers all BHP-eligible consumers through the state’s longstanding “MinnesotaCare” program, now operating under a Medicaid waiver—serve numerous BHP-eligible consumers outside the individual market. No risk-adjustment system or comparable data-gathering mechanism allows a prospective comparison between the average risk level of such consumers and those who will enroll in the 2014 individual market. As a result, CMS is allowing states, for the 2015 BHP program year, to have the PHF determined retrospectively, after the conclusion of the 2015 BHP program year. A state choosing this option must, by August 1, 2014, have proposed a protocol to CMS for gathering the information needed to determine the PHF. CMS must approve the protocol by December 31, 2014. If information gathered through the protocol shows the need to change CMS’ 2015 payments, adjustments would be carried out through increases or reductions to the state’s later BHP payments.
Links to such rates are available through the map at http://www.cdc.gov/tobacco/widgets/index.htm#widget.
Estimating Federal BHP Funding Levels
Most state individual markets, including QHPs, use HHS’s default ratios between premiums charged to adults age 21-24 and individuals of other ages. If more than one person within a family enrolls in a plan, the family premium combines the premiums charged to each family member, based on their ages. A few states—the District of Columbia, Massachusetts, Minnesota, New Jersey, and Utah—depart from the HHS default ratios in varying premium charges based on age. These states still determine family premiums based on the combined age-specific premiums charged to each enrolling family member. For HHS’s default ratios and the ratios used by the latter states, see Center for Consumer Information and Insurance Oversight (CCIIO). State Specific Age Curve Variations.August 9, 2013. http://www.cms.gov/CCIIO/Programs-and-Initiatives/Health-Insurance-Market-Reforms/Downloads/state-specific-age-curve-variations-08-09-2013.pdf. New York and Vermont do not permit premiums to vary based on age. Premiums vary based on family enrollment, depending on the characteristics of the enrolling family. In each state, coverage for two adults costs twice as much as coverage for one adult. For information about how premiums change when children are involved, see CCIIO, “State Specific Family Tier Ratios,” Market Rating Reforms: State Specific Rating Variations. Updated: July 11, 2014. http://www.cms.gov/CCIIO/Programs-and-Initiatives/Health-Insurance-Market-Reforms/state-rating.html#age.
The purpose of this example is to illustrate our suggested approach to calculating federal BHP payments, not to provide up-to-the-minute, accurate estimates for Washington State. After this example was developed, final QHP premiums for 2015 were announced. To obtain more accurate and current estimates, Washington State officials and stakeholders would need to revise these calculations using actual 2015 premiums, rather than the projections we developed based on state insurance officials’ analysis.
Dekker Dirksen, Community Health Plan of Washington/Community Health Network of Washington, personal communication, July 2014.
Washington State Health Benefits Exchange, April 23, 2014. Health Coverage Enrollment Report: October 1, 2013 - March 31, 2014. http://wahbexchange.org/files/2713/9888/1218/WAHBE_End_of_Open_Enrollment_Data_Report_FINAL.pdf.
To derive the weighted average, we first multiple the premium in a county by the number of QHP enrollees in that county. For example, we multiply $221.14 in Adams County by the 451 QHP enrollees and obtain a product of $99,734.14. We combine such county-specific products for all counties, which equals $34,028,555.85, and divide by the total number of QHP enrollees statewide, which is 152,690. The resulting average is $222.86.
Mike Kreidler, Washington state Insurance Commissioner, “Seventeen health insurers file more than 230 plans for 2015 - average proposed rate change 8%,” News Release, May 13, 2014, http://insurance.wa.gov/about-oic/news-media/news-releases/2014/5-13-2014.html; Jeffrey Naas, Washington State Office of the Insurance Commissioner, personal communication, July 2014.
For information about each individual state’s approach to age rating, see CCIIO, Market Rating Reforms: State Specific Rating Variations, Updated July 11, 2014, http://www.cms.gov/CCIIO/Programs-and-Initiatives/Health-Insurance-Market-Reforms/state-rating.html; CCIIO, State Specific Age Curve Variations, August 9, 2013, http://www.cms.gov/CCIIO/Programs-and-Initiatives/Health-Insurance-Market-Reforms/Downloads/state-specific-age-curve-variations-08-09-2013.pdf.
Very few BHP-eligible consumers live in households with more than five members.
For BHP-eligible consumers under 138% FPL, sample size considerations prevented us from developing eligibility estimates within the smaller FPL ranges used by the federal payment methodology (0-50, 51-100, and 101-138% FPL). We assumed an even distribution of BHP enrollees by FPL income levels from 0 to 138% FPL, as provided by the federal payment methodology for narrower FPL ranges. If BHP-eligible consumers are unevenly distributed among the three federal-specified FPL ranges below 138% FPL, our estimated payment amounts for consumers under 138% FPL may be inaccurate. However, given the relatively small size of the under-138%-FPL population among BHP-eligible consumers, the impact on calculating a state’s average federal payment per BHP-eligible consumer is likely to be modest.
Hawaii and Alaska would be treated differently, because the FPL equals different income amounts in those states than in other states.
The only exceptions are fully community-rated states, where family premiums vary based on the number of adults and children enrolled in coverage. CCIIO, Market Rating Reforms: State Specific Rating Variations.
Put differently, the two BHP-eligible members receive family coverage for which a premium of $850.46 is charged, the household payment requirement is $106.30, and the household PTC is $744.16—precisely twice the $372.08 received by each BHP-eligible member.
Premera Blue Cross, “Individual Filing – Effective 1/1/2014,” Exhibit 6.2 in Exchange Rates – Silver Plans; LifeWise Health Plan of Washington, “Individual Filing – Effective 1/1/2014,” Exhibit 6.2 in Exchange Rates – Silver Plans.
Group Health Cooperative, 1/1/2014 Individual Rate Filing, “Exhibit 11 - Final Rates.”
These counties are Benton, Franklin, King, Walla Walla, and Yakima. Dirksen, op cit. For the distribution of enrollees by County, see Table 2.
These numbers come from Tobacco Control State Highlights 2012, http://www.cdc.gov/tobacco/data_statistics/state_data/state_highlights/2012/zip_files/highlights.zip.
As explained by CMS in its proposed 2016 BHP methodology, “For the BHP payment rate cell for persons ages 21-34, we would calculate the factor as (4/14 * the utilization rate of 18-24 year olds) plus (10/14 * the utilization rate of 25-44 year olds), which would be the weighted average of tobacco usage for persons 21-34 assuming a uniform distribution of ages; for all other age ranges used for the rate cells, we would use the age range in the CDC data in which the BHP payment rate cell age range is contained.” CMS. Basic Health Program: Federal Funding Methodology for Program Year 2016.
CMS, Basic Health Program: Federal Funding Methodology for Program Year 2016.
Conclusion: placing federal payment estimates in context
Dorn and Tolbert 2014.
Appendix: The Characteristics of BHP Eligibles by State
These estimates will be almost the same as those for 2015, with very small changes reflecting population growth. A state analyzing BHP implementation for 2015 could use the tables in this appendix to develop the kind of fiscal analysis described in the body of this report.
Among consumers with incomes between 139 and 400% FPL who are offered ESI, between 97% and 99.8% of such offers meet the ACA’s definition of affordability. Even among consumers in this income range who do not accept ESI offers, between 87% and 99% of the rejected offers are affordable. See the U.S. panel in table 1 in Matthew Buettgens, Stan Dorn, Habib Moody. Access to Employer-Sponsored Insurance and Subsidy Eligibility in Health Benefits Exchanges: Two Data-Based Approaches. Washington, DC: Urban Institute (prepared for the California HealthCare Foundation), Dec. 2012, http://www.urban.org/UploadedPDF/412721-Access-to-Employer-Sponsored-Insurance.pdf.
See Buettgens, Dorn and Moody, 2012.
Passel, J. and D. Cohen. 2009. “A Portrait of Unauthorized Immigrants in the United States.” Washington, DC: Pew Hispanic Center.
Treasury Inspector General for Tax Administration, “The Internal Revenue Service Needs a Coordinated National Strategy to Better Address an Estimated $30 Billion Tax Gap Due to Non-filers,” November 2005, Reference Number 2006-30-006.
“Internal Revenue Service Data Book 2003,” Internal Revenue Service, 2003.
Plueger, D, “Earned Income Tax Credit Participation Rate for Tax Year 2005,” Internal Revenue Service, 2009.
For households with 1 BHP-eligible member, groups would include 5 age ranges, 4 FPL income ranges, and 5 household sizes (with households of 5 or more members constituting the largest household), for a total of 100 groups (5x4x5=100). Households with 2 BHP-eligible members have the same number of age and FPL income ranges, but only 4 household sizes, since a 1-person household cannot have 2 BHP-eligible members. Accordingly, this second set includes 80 groups (5x4x4=80). The final set, consisting of households with 3+ BHP-eligible members, has only 3 household sizes, so it includes 60 groups (5x4x3=60). Altogether, these three sets include 240 groups (100+80+60=240).
Specifically, we reweighted by minimizing cross-entropy. Martin Wittenberg, An introduction to maximum entropy and minimum cross-entropy using Stata, The Stata Journal (2010) 10, Number 3, pp. 315-330.