Cost of the Newly Insured Under the Affordable Care Act

The Society of Actuaries is pleased to make available draft material examining the cost of the newly insured under the Affordable Care Act. The draft material is being exposed for a comment period before the final material is issued. The deadline for comments is January 25, 2013. The material was authored by Randy Haught of Lewin Inc. and John Ahrens of Optum Inc.

Health practitioners and other interested parties are encouraged to provide comments on the draft material. Comments may be directed to Steve Siegel, SOA Research Actuary at or by leaving comments below.

Thanks for your interest!

Sharing is caring.
  • Subscribe to our feed
  • Tweet about this post
  • Share this post on Facebook
  • Share this post on Google
  • Share this post on LinkedIn


One response to "Cost of the Newly Insured Under the Affordable Care Act"

  • Stephen Arnhold says:

    Humana Comments
    Prepared by Stephen Arnhold, FSA, MAAA, CFA (with input / review from Humana’s actuarial community) January 25, 2013

    These comments are in response to the “EXPOSURE DRAFT: Cost of the Future Newly Insured under the Affordable Care Act (ACA). Our review of the written document revealed many assumptions and results that we don’t believe produce plausible outcomes. Since the modeling is highly interdependent, this causes us to question the validity of the final results and conclusions, particularly when broken down at a state or demographic cell level. We were not able to review all of the individual state results in the Excel spreadsheet due to time constraints in the request for comments. This should also not be considered a comprehensive or complete review, but we wanted to provide some direct, objective feedback. Without access to the underlying models a technical review is not possible. We believe the model and paper should be revised (based on the comments below) and is not yet ready for publication by the Society of Actuaries.
    1) Figure 1: Changes in Sources of Coverage under the ACA for Wisconsin in 2014 (page 11), contains some questionable transitions in coverage, particularly in relation to small group coverage:
    a. 25.8% (174,937/678,829) of the existing employer 2-50 membership moves to the Employer Exchange. That percentage of membership on the Exchange seems unusually high in relation to other models. The distribution model for small employer coverage has historically been broker driven and there is minimal incentive for small groups to choose Exchange coverage. (There is a potential small business tax credit for employers of 25 employees or less, where average income is below $50,000; our modeling suggests less than 10% of small employer will actually qualify for this tax credit).
    b. 14.4% of the currently uninsured move to group coverage. That is a very high percentage given that group coverage will not necessarily be more affordable. (That assumption may be plausible, but only if the state of the economy improves markedly, with unemployment rates approaching zero). The underlying assumptions need to be a combination of more 50+ groups offering coverage due to the penalty and more individuals electing coverage due to the mandate. The vast majority of large groups already offer coverage. Small groups have shown no historic inclination to offer coverage for small incentives such as temporary tax incentive and there is no mandate for them to do so. (Also, the vast majority of the uninsured make less than 400% of FPL, so subsidies would be available to them by purchasing on exchange).
    2) Figure 2: Changes in Sources of Coverage under the ACA for Currently Uninsured by Age, Income and Self-Reported Health for Wisconsin in 2014 (page 14):
    a. The 19-24 age group has a significant percentage of 21.9% going to employer Exchange and Private Employer coverage. The requirement to continue dependent coverage to age 26 has been in place since 2010. There does not appear to be anything to motivate this migration since dependents are already required to be covered.
    b. The Below 138% FPL group has a significant percentage of 11.9% going to employer Exchange and Private Employer coverage. Where does the incentive for individuals that are presumptively eligible for Medicaid come from the obtain group coverage? There is clearly a greater cost sharing element to just about every group coverage rather than Medicaid, both in the form of employee contribution and cost sharing.
    3) Figure 3: Number and Cost of Newly Insured by Age, Income and Self-Reported Health Status in Wisconsin in 2014 (page 15)
    a. The Average Monthly Cost Post-ACA is highest for the 45-54 Age group at $786. That cost is higher than the 55 & over age group at $730. We have never seen an age curve where morbidity decreased at the highest ages, including the ACA age curve which has a sharp increase. The percentage increase for this age group is also disproportionately high at 254.9%, over twice that of any other age group.
    b. The 138%-199% FPL segment has the lowest Post-ACA claims cost at $243 and the lowest Percent Change in Claim Costs at 68.7%. This segment has the lowest cost sharing and presumably the highest pent-up demand (document lists as out of scope in the following section, but we believe this could persist for a significant period of time), this outcome doesn’t seem plausible. This segment is also expected to have high turnover which could further raise costs.
    4) Figure 6: Distribution of Non-Group Coverage Pre- and Post-ACA by age, income and health status in Wisconsin in 2014 (page 19)
    a) The 138%-200% FPL cohort shows a significant decrease in costs of 18.9%, while every other segment experiences a significant increase in costs.
    b) The variance in costs between the 200%-300% FPL cohort and the other income levels is extreme, $498 Average Monthly Cost versus $340 for the next lower income group and $337 for the next higher group.
    c) The 400% FPL and over group has extremely high costs in relation to the other cohorts and the biggest percentage change of 83.1%. Yet this group has the lowest increase in Number and most closely represents the current Non-Group market.
    5) Figure 9: Distribution of Medicaid Enrollees Pre- and Post-ACA by Age, Income and Health Status in Wisconsin in 2014 (page 23)
    It is unclear why the number of individuals covered by Medicaid in each FPL cohort above 138% FPL decreases. There is nothing in the law that decreases Medicaid eligibility for these groups which generally fall into one or more special eligibility groups.
    6) Figure 13: Average Costs for Members that Leave Employer Coverage and How They Sort Into Programs under the ACA in Wisconsin in 2014
    a. There is extreme variation in the Average Costs based on the Group Sizes they are coming from. The 500-999 Group Size going to Medicaid is $187 and the 1000-4999 group size is $608. The wide variation in these cost figures is not reasonable or explainable.
    b. There is extreme variation in the Average costs based on the destination. The 500-999 Group Size going to Medicaid is $187, to Non-Group is $1,030 and to Uninsured is $571. This is not a logical pattern shared across group sizes, the corresponding numbers for the 1000-4999 Group Size are $608, $536 and $224.
    7) Decision for Uninsured to take Non-Group Coverage (page A-9)
    We believe the price elasticity of 3.4 is extremely high. A 3.4% increase in the number of covered individuals could be a cross-elasticity if the price is close to the prevailing market price, but the assumption that a 1% decrease in price will drive a 3.4% increase in the size of the market is inconsistent with any data we have seen in either the individual or group markets, over the last 20 years.
    8) Figure A-5: Percentage of People with Non-Group Insurance who Discontinue Coverage (page A-10)
    The assumption that no one with expected claims above $10,000 drops coverage when there is a rate increase in excess of 50% seems unrealistic. Some people will have an affordability issue with coverage at that level of rate increase and will drop coverage regardless of their expected claims.
    9) Figure A-6: Individual Decision to Purchase Coverage through the Exchange (page A-11)
    a. The assumption that 0% of the current non-group market purchase through the exchange is not realistic, as some individuals will use the exchange as a quoting and purchasing mechanism. Many will be subsidy eligible, but only via purchase on the exchange.
    b. The assumption that 100% of the new purchases without subsidies purchase on the exchange is unrealistic. Some individuals will seek out existing agents or respond to carrier and broker advertising. In Massachusetts, the vast majority of purchases are off exchange, where carriers offer broader networks and more flexible plan options.
    10) Figure A-7: Uninsured Workers Who Have Declined Employer Coverage under Current Law Who Take That Coverage as a Result of the Mandate (A-12)
    The patterns in this table don’t make sense. Fewer workers should take the coverage as rates increase and more workers should take the coverage as rates decrease. It is entirely possible that the size of the rate increase could make this factor be negative; that is more workers drop than add coverage due to affordability.
    11) Figure A-8 Employers Who Decide to Offer Coverage Due to Price Changes by Change in Premiums and Group Size (page A-13)
    The assumption that small employers start offering coverage based on a small decrease in cost of coverage seems questionable. Small employer coverage has consistently decreased over an extended period of time. Previous attempts at temporary tax or price subsidies have not been shown to drive a significant increase in coverage offered. The correct assumption here should be that small groups are more likely to drop coverage than to start offering coverage for the same price change in opposite directions.
    12) Figure A-9: Employer Decision to Start Offering Coverage Due to Increased Worker Demand for Coverage (worker weighted) (page A-14)
    It seems questionable that groups that are not subject to a mandate will begin to offer coverage, especially when workers are in many cases eligible for subsidized coverage and in some cases Medicaid.
    13) Figure A-12: Employer Decision to Offer Coverage in the Exchange
    Why would 45% of firms move their coverage to the Exchange? Early indications are that only a subset of carriers/plans will be offered on the SHOP exchanges. Firms that currently have coverage utilize brokers almost exclusively to procure that coverage and private exchanges are being developed as an alternative to the SHOP exchanges. It is highly questionable that almost half of small firms would move to the SHOP in a relatively short period of time.
    14) Figure B-4: Age Rating by Single-year of Age
    a. At the youngest ages, the Female family rates seem disproportionately high and the male rates seem correspondingly low. At age 17, male headed families have a relatively of .4016 and female headed families are at 1.6568. This 1:4 relatively quickly reduced to less than 1:1.5 by age 19.
    b. At the highest ages, the female family rates seem unusually high. At age 64, the female family factor is 3.6889 and the male factor is 2.1359. At age 60 the male factor is higher.
    15) Figure 1 of the state tables shows that around 15% of the uninsured will be enrolling in employer coverage compared to 20% in individual/non-employer. The 15% seems quite high.
    16) Figure 1A of the state tables shows that the current morbidity for the uninsured segment is about half of that of the non-group market (excluding high-risk pool) pre-ACA. This morbidity level for the uninsured market seems quite low.
    17) Figures 5 and 1A of the state tables show that the morbidity for those individuals migrating from the current employer market to the non-employer market is about 32% higher than the overall morbidity for the current employer market pre-ACA. That 32% varies based on employer size and ranges from approximately 9% for the 2-50 employee size up to 54% for the 1001+. The net result of these assumptions combined with those of the previous comment is that individuals migrating from the employer market to the non-employer market account for 9x more of the overall morbidity increase for the non-employer market relative to the uninsured. This does not seem plausible. Please note that the numbers quoted here represent an average of about 10 states.
    18) Figure 2 of the state tables shows that approximately 98% of the non-employer market that is 400%+ FPL will be on-exchange. That does not seem plausible as there is no incentive to be on exchange for that segment although there are incentives to not be on-exchange including the use of brokers.
    19) Figure 2 supports a higher income level for on-exchange non-employer population than off-exchange non-employer population. This does not seem plausible since there are significant incentives for lower income individuals to enroll on-exchange.
    Respectfully submitted on behalf of Humana,
    Stephen Arnhold, FSA, MAAA, CFA,
    Brandon Combs, FSA, MAAA,
    Joyce Bohl, ASA, MAAA,
    January 25, 2013

Leave a Comment