JOli N A. KI1ZHAB[R, MD Governor

Transcription

1 July 31, 2012 JOli N A. KI1ZHAB[R, MD Governor The Affordable Care Act will provide access to insurance coverage for thousands of Oregonians who need it most. In 2014, Oregonians with pre-existing health conditions will no longer be denied coverage, and those with coverage will have better access to preventive care, prescription drug coverage and mental health treatment. The attached report estimates the financial impact of increased health care access and essential benefit plans on the cost of insurance, and is a tool to help the State of Oregon develop strategies to mitigate adverse impacts on some individuals and small businesses. Three points of context are important when considering the report's findings: I. The report is a snapshot of reform to the current health care financing and delivery system; it does not account for cost savings Oregon anticipates from its health transformation initiative. Indeed, the report's findings point to how important it is to follow through on progress to date to change the health care delivery system to make it more affordable. We need to learn from the implementation of Coordinated Care Organizations (CCOs) and begin plans to incorporate the concept into the commercial insurance market. Eliminating waste and inefficiencies, and driving down health care costs by integrating care, reducing hospitalizations and reducing expensive and unnecessary procedures will be critically important as we expand health care access. 2. Better access to health insurance coverage will reduce the number of people using expensive hospital emergency departments for primary care with consequent reductions in uncompensated care. We should expect this to lead to more timely, efficient and coordinated care for individuals and lower costs to the system. 3. Approximately half of those who already have insurance in the individual market and nearly 75 percent of those who are now uninsured will receive financial assistance through tax credits and cost sharing incentives that will lower the costs of coverage they purchase through the Oregon Health Insurance Exchange. Finally, the projected cost increases in the individual insurance market reflect the true cost of our current health care system - costs that exist today but are hidden. These include costs the system bears for the uninsured who seek care in hospital emergency rooms or who postpone care until absolutely necessary and for high-risk individuals who are covered through a state program and whose high costs are passed on in the premiums others pay throughout the insured market. This increased transparency can help us better understand and address the drivers of those costs as we work to improve access to care and the health of all Oregonians. While this report focuses on the cost of expanding coverage and providing better benefits to Oregon's citizens, our focus will continue to be getting the best value and quality for our health care dollars. It is not just about coverage, it is about health. Sincerely, 254 STATE CAPITOL, SALEM OR (503) FAX (503)

2 Actuarial Analysis: Impact of the Affordable Care Act (ACA) on Small Group and Individual Market Premiums in Oregon Prepared on July 31, 2012 Prepared for: The Prepared by: Wakely Consulting Group Julie Peper, FSA, MAAA Luke Rodgers, ASA, MAAA Julia Lambert, FSA, MAAA Kelsey Stevens, FSA, MAAA Ross Winkelman, FSA, MAAA Jon Kingsdale, Ph.D Pyramid Court Suite 260 Englewood, CO Tel Fax

3 TABLE OF CONTENTS 1. EXECUTIVE SUMMARY ANALYSIS OF THE OREGON INDIVIDUAL MARKET Summary of Individual Analysis Current Regulations and Market Composition Data Received ACA Impact on Individual Market Premium Tax Credits and Cost Sharing Subsidies Additional Requirements and Considerations How to Reduce Churning between the Individual Exchange and Medicaid? ANALYSIS OF THE OREGON SMALL GROUP MARKET Summary of Small Group Analysis Current Rating and Underwriting Rules in Oregon Current Coverage and Costs in the Oregon Small Group Market ACA Impact on Small Group Market Summary INDIVIDUAL AND SMALL GROUP MERGER IMPORTANT CAVEATS APPENDIX A ACA IMPACT ON PORTABILITY APPENDIX B ACA IMPACT ON OMIP/FMIP APPENDIX C ACA IMPACT ON HEALTHY KIDSCONNECT July 31, 2012

4 1. EXECUTIVE SUMMARY Wakely was retained by the state of Oregon to analyze the impact of the Affordable Care Act (ACA) on Oregon s individual and small group markets in This report presents the results of Wakely s work. All estimates presented in this report are specific to the state of Oregon and will likely be different for other states. The following components are discussed in this report: 1. Current Oregon regulations and market composition for the individual (non-group) and small group markets. 2. Analysis of the impact of the ACA reforms on the various Oregon markets, including the impact on benefit plan offerings, rating and underwriting, as well as the impact due to newly insured individuals and premium tax credits and cost sharing subsidies under ACA. This analysis was completed for the following markets: a. Individual b. Portability c. Oregon Medical Insurance Pool (OMIP) and Federal Medical Insurance Pool (FMIP) d. Healthy KidsConnect (HKC) e. Small Group f. Small Groups currently in associations 3. Impact of merging the various markets: a. Individual submarkets, which include the current individual market, OMIP/FMIP, portability and HKC. b. Small group submarkets, which include the current small group market and small groups currently in associations. c. Resulting new individual and small group markets. 4. Churning between the individual exchange and Medicaid. 5. Impact of increasing small group to 100 members. For the individual, portability, OMIP/FMIP, Healthy KidsConnect and small group markets, we received data from the largest eight insurers in the state. The data received includes: summarized plan benefit packages, premiums, claims, underwriting, non-benefit expenses, commissions, and current members demographic information. We supplemented this information with publicly available rate filings and survey information, data from other states, and information provided by the Oregon Department of Consumer and Business Services, Insurance Division (the Department). We reviewed this information July 31, 2012 Page 1

5 for reasonability, but did not audit the information. Any errors in this data may materially impact the results of our analysis. The following describes key results of the analysis. The full report should be referenced to fully understand the assumptions, approach and results. Individual 1. Wakely estimates that premiums for a combined individual, portability, OMIP/FMIP and HKC market will increase by approximately 24% (on average) due to the ACA changes, before premium tax credits are considered. However, approximately 7% of this increase will result in lower out of pocket benefit and cost sharing payments. In addition, the ACA offers premium tax credits and cost sharing subsidies. Therefore, Wakely s best estimate of ACA changes is a 23% decrease in individual market member out of pocket costs (premiums plus cost sharing). The table directly below details the premium impact to the average individual/portability/omip/ FMIP/HKC member. Table 1: Individual Market Average Member Impact on Out of Pocket Costs Average Member Impact 2014 Best Estimate ACA Requirements 24% Premium Tax Credits -27% Net Premium Impact -10% Out of Pocket Reductions Essential Health Benefits -5% Minimum Bronze -2% Cost Sharing Subsidies -8% Total Out of Pocket Reductions -14% Overall Change in Member Costs -23% 2. Premium tax credits will offer premium relief to approximately half of the currently insured market and almost 75% of the uninsured who are expected to enroll in the individual market in Before consideration of premium tax credits, Wakely estimates that average individual market premiums will increase by 27% to 55%. This increase is comprised of 2% to 30% due to ACA requirements and another 20% to 24% due to the addition of portability, high risk pool, and Healthy KidsConnect members into the individual market. Wakely estimates that out of pocket costs will decrease by 6% to 11% due to increased benefit and cost sharing coverage, prior to July 31, 2012 Page 2

6 any cost sharing subsidies. This results in a net average estimated cost change to the average member of 20% to 40% prior to any premium tax credits or cost sharing subsidies. 4. The impact of the ACA changes will vary greatly by individual enrollee. It is difficult to predict how these changes in premium rates will affect individual's choices regarding coverage options. Small Group 1. Wakely estimates that the range of average small group premium changes will be from a 5% decrease to a 16% increase due to the ACA reforms and merging of small groups currently in associations. Most of the uncertainty in this range is due to the inherent difficulty in estimating the number of healthy groups that may leave the market because of larger rate increases. 2. The impact of the ACA changes will vary significantly by group. The groups receiving the largest increases will have a disproportionate share of young, healthy employees. Individual / Small Group Merger 1. If the individual and small group markets are merged, Wakely estimates that premiums for the new small group market would decrease by approximately 4% and premiums for the new individual market would increase by approximately 4%. Because the ACA will increase the number of insured people, hospitals and physicians will be less likely to shift costs from individuals who cannot pay ( uncompensated care ) to the privately insured. The ACA also reduces federal payments available to hospitals for the Disproportionate Share Hospital (DSH) program. Therefore, the DSH program reductions provide a somewhat offsetting impact to the increase in insured individuals and groups. However, the overall expected impact of the ACA is an increase in provider funding. This increase may result in a decrease in provider payment rates, since some providers may not need to subsidize uncompensated care to the same degree they have historically. Results for each component of the analysis are included below. Please see individual sections of the report for descriptions of our methods, assumptions, data and inherent limitations with our estimates. Estimates of the impact of healthcare reform provisions are inherently uncertain because of the large number of forces affecting the insurance market, including actions by consumers and health plans. Also, many structural decisions regarding the exchange and the insurance market have not been decided by the state and federal regulations and guidance are still pending; such decisions, regulations and guidance may significantly impact premiums and product offerings. A summary of the results and conclusions is listed below. Detailed results and discussion begin in Section 2 of this report. 1.1 Overview - Individual Market under ACA Under the ACA the current individual market, OMIP/FMIP, portability, and Healthy KidsConnect are expected to comprise the new individual insurance market. Throughout the report the impact of the July 31, 2012 Page 3

7 ACA requirements to each submarket will be separately discussed, followed by the impact of merging the submarkets into one larger market. Individual Market Estimated changes to individual market premiums due to ACA regulatory reforms, as compared to the current Oregon market (as of 2011) are shown in Table 2 below. Each of the ACA changes included in Table 2 is discussed in detail in this report. Note that Table 2 below represents the estimated 2014 impact; Table 3 provides a comparison of best estimates, by year, from 2014 to As shown in Table 3, the estimated impact of some of these changes, such as reinsurance, vary considerably by year. The best estimate of the increase in premiums is approximately 38% above what it would be without the ACA changes. This reflects the change in the premium required to cover the health risks of the expected population after the ACA changes and changes in costs including benefit changes required under the ACA. It can also be viewed as the expected change in the filed rates of an insurer. The following table shows the components of this change: Table 2: Individual Market Premium Impacts under ACA (2014 compared to 2011) Individual Premium Impacts Average Premium Impact ACA Requirement Low Best Estimate High Essential Benefits Requirement 5% 6% 7% Bronze Minimum Act. Value (includes Max OOP limit) 1% 2% 3% Minimum Loss Ratio = 80% -2% -1% 0% Morbidity Change (due to new insured/uninsured) 10% 15% 25% Age Slope Limited to 3:1 0% 0% 0% Provider Fee 1% 1% 1% Reinsurance Program -9% -8% -7% Elimination of OMIP Assessment -2% -1% 0% Subtotal (ACA Requirements) 2% 13% 30% Individual Submarket Merger 24% 22% 20% Total Premium Impact 27% 38% 55% As shown above, we estimate that average individual market premiums will increase significantly due to the ACA reforms and the influx of newly insured individuals. The increase in premiums is mostly driven by benefit increases and the morbidity (or population risk) changes from newly insured individuals entering the market. Some of the impacts, such as age factors, have no overall expected impact but the impact to a particular insured can be significant. The impact from merging the individual submarkets is July 31, 2012 Page 4

8 also significant; this impact is discussed below in the subsection titled Combining the Individual Submarkets. The component Morbidity Change (due to new insured / uninsured) in the table above reflects the average risk change to the current individual market due to the migration of some members out of the individual market and into Medicaid, as well as migration from the uninsured and small group market into the individual market. The impact of the risk corridor provision is not included in the estimates above. While the risk corridor provision will clearly affect plans and insurers differently, it is impossible to predict this impact in advance. While many of the changes increase premiums, they also decrease out of pocket expenses for individuals. For example, while the essential benefits requirement is expected to increase premiums by 6%, this increase will be offset by a decrease in out of pocket expenses as these additional benefits will be covered post-aca. Excluding changes that decrease out of pocket expenses, since these have offsetting effects, results in a range of overall health cost impacts of 20% to 40%, with a best estimate of 27%. This is prior to any further relief due to premium tax credits and cost sharing subsidies. Table 3: Individual Market Premium Impacts under ACA ( ) Individual Premium Impacts Average Premium Impact ACA Requirement Essential Benefits Requirement 6% 6% 6% Bronze Minimum Act. Value (includes Max OOP limit) 2% 2% 2% Minimum Loss Ratio = 80% -1% -1% -1% Morbidity Change (due to new insured/uninsured) 15% 15% 15% Age Slope Limited to 3:1 0% 0% 0% Provider Fee 1% 1% 1% Reinsurance Program -8% -4% -2% Elimination of OMIP Assessment -1% -1% -1% Subtotal (ACA Requirements) 13% 19% 21% Individual Submarket Merger 22% 20% 18% Total 38% 42% 42% As shown in Table 3, the estimated impact of reinsurance varies significantly by year, from approximately -8% in 2014 to an estimated -2% in The provider fee also varies by year but by a small amount that is not seen due to rounding. July 31, 2012 Page 5

9 Portability Coverage through the portability market is relatively rich. The ACA requirements will not impact members in this submarket as much as it does members in the current individual market. Wakely estimates that 2014 portability premiums will increase from 42% to 59% due to ACA changes, with a best estimate of a 50% increase (excludes the impact of any premium tax credits). The primary driver of the estimated premium changes is the current subsidization of portability premiums. Since the portability market is subsidized and many insurers experience loss ratios above 100%, these members would likely experience premium increases due to more appropriate pricing. However, the submarket merger (discussed below) is estimated to decrease premiums to portability members by a best estimate of 39%, which offsets these increases. After the submarket merger, portability members are estimated to have an overall premium impact of an 8% decrease. The change in the age slope will not impact average premiums significantly, but will have a large impact to some individual members. Current age slopes vary by insurer, with one insurer varying rates by gender. Most insurers use premium age slopes that vary by less than the 3:1 ACA requirements for portability coverage. Assuming these members join the individual market starting in 2014 which will be subject to a 3:1 age ratio limit, many portability members will see premium impacts opposite of what the current individual market will experience (outside of other changes, older portability members will see a rate increase while younger ones will see a decrease due to ACA age ratio limits). Oregon Medical Insurance Pool/Federal Medical Insurance Pool Similar to the portability market, OMIP/FMIP plans have relatively rich benefits and cost sharing. Therefore, impacts due to ACA requirements are nominal. The impact of ACA regulatory reforms on OMIP/FMIP premiums is estimated between +100% and +159%, with a best estimate of +141%. Elimination of the OMIP/FMIP assessment is the primary driver of the estimated premium changes with potential offset due to the submarket merger. The submarket merger is estimated to decrease OMIP/FMIP premiums by 67%, for a resulting overall premium decrease between 12% and 38% with a best estimate 21% decrease. Also similar to the portability market, the change in the age slope will likely not have an overall premium impact, but expected changes in the age slope under the ACA will have the largest member impact. Current age slopes for OMIP/FMIP premiums are close to 2:1, so the expansion to an assumed 3:1 in a combined individual market will decrease rates for the younger members and increase rates for the older members. Healthy KidsConnect Market It is not known if the HKC members will migrate to the Medicaid or individual market (likely some of each). For discussion purposes, we have assumed they will move to the new individual market. July 31, 2012 Page 6

10 Healthy KidsConnect premiums are expected to decrease between 12% and 5% due to the ACA changes, with a best estimate of a 9% decrease. Overall, most of the regulatory requirements should have very little impact on average premiums. The most impactful requirements are reinsurance (especially in 2014) and the minimum loss ratio. HKC benefits encompass almost all of the essential health benefits requirements, including pediatric dental. Age slope requirements should not impact this market. The submarket merger is estimated to have minimal impact on the HKC market resulting in an overall premium decrease of 14% to an increase of 3% with a best estimate of a 5% decrease. Combining the Individual Submarkets The merging of the current individual market with the portability, OMIP/FMIP and Healthy KidsConnect markets will create a larger individual market; hereinafter, this combined market will be referred to as the new individual market. To understand the impact of merging these markets we compared the normalized allowed (before cost sharing) costs for the various submarkets, all adjusted for the impact of ACA requirements, including the migration from the uninsured into the current individual market. Costs were further normalized to account for any differences in age, geographic concentrations and other rating variables. We estimate that merging these submarkets into the new individual market will change post-aca premiums as follows: 1. the current individual market will increase by 22%, 2. post-aca portability premiums will decrease by 39%, 3. post-aca OMIP/FMIP premiums will decrease by 67% 1, and 4. Healthy KidsConnect premiums will increase by 4%. A shown above in Table 2, combining the merging of the individual submarkets with the impact of ACA requirements results in the following (2014) post-aca premium changes: the current individual market will increase by 38%, post-aca portability premiums will decrease by 8%, post-aca OMIP/FMIP premiums will decrease by 21% and Healthy KidsConnect premiums will decrease by 5%. Premium Tax Credits and Cost Sharing Subsidies Beginning in 2014, some lower income individuals and families will be eligible to receive premium tax credits and cost sharing subsidies to make health insurance more affordable. It is estimated that the 1 Since OMIP/FMIP and portability premiums are currently subsidized, the premium impact shown is related to the full premiums that would need to be charged to cover expenses rather than the impact to the currently subsidized premiums. July 31, 2012 Page 7

11 premium tax credit subsidies would result in an average premium decrease of 10% for the combined individual market. This impact is based on the following assumptions: 1. Best estimate assumptions; 2. The new individual market which includes the current individual market, Portability, OMIP/FMIP, and Health KidsConnect; 3. An estimate for the second lowest silver plan premium; and 4. An assumption that individuals and families with incomes less than or equal to 133% of the Federal Poverty Level (FPL) will not be enrolled in the health insurance exchange. The impact to any one member or family is significantly affected by the income of that member or family. Therefore, while premiums are expected to decrease 10% on average, the premium impact on any given member or family will vary from this value, potentially significantly. While our primary focus is on the premium change to the currently enrolled members, it should be noted that the average premium and cost sharing subsidies vary significantly when looking at the currently enrolled and the newly enrolled. The newly enrolled population is expected to have lower incomes and thus receive higher premium tax credits (on average). Individuals who qualify for premium credits and are enrolled in a silver plan in the exchange will also be eligible for assistance in paying their cost sharing. Using similar assumptions as those listed in this section (above), it is estimated that the cost sharing subsidies will decrease the cost sharing in a silver level plan by an average of 23%, which translates to an estimated 8% decrease to average premium. 1.2 Overview - Small Group Market under ACA Small Group Market Table 4 below includes estimates of ACA changes to small group market premiums (on a PMPM basis) beyond current legislation in-force in Oregon. The impact of the ACA requirements that have already gone into effect, such as the dependent definition expansion to age 26, are not included in the table below. The impacts below do not vary significantly from 2014 to 2016 with the annually decreasing reinsurance assessment the key driver of any differences. July 31, 2012 Page 8

12 Table 4: Changes to Small Group Market Premiums under ACA (2014 compared to 2011) Average Premium Impact Range of Group Impact ACA Requirement Low Best Estimate High Low High Compliance with MOOP, deductible 0% 0% 0% Varies Varies Essential Benefits Requirement 1% 2% 3% 1% 3% Bronze Minimum Act. Value 0% 0% 1% 0% 6% Removal of OMIP and Portability Assessment -3% -2% -1% -3% -1% Minimum Loss Ratio = 80% 0% 0% 0% -6% 0% Age / Gender Slope 1, 5 0% 0% 0% -10% 50% U/W - Participation 1,2 0% 0% 0% -14% 8% U/W - Contribution 1,3 0% 0% 0% -7% 4% U/W - Health Status 4 0% 1% 3% -5% 5% Morbidity Change -5% 0% 5% N/A N/A Provider Fee 0% 1% 1% 0% 1% Reinsurance Assessment 1% 1% 1% 1% 1% Subtotal (ACA Requirements) -5% 3% 14% -37% 95% Submarket Merger 0% 1% 2% 0% 2% Total -5% 4% 16% -37% 99% 1 Premium changes due to the potential impact of selection are not included in this analysis. 2 Two carriers were unable to provide distribution data. They have been excluded from this analysis, though results could be material. Additionally, one carrier does not consider employee participation in rating. 3 Four carriers do not consider employer contribution in rating. 4 Two carriers were unable to provide data. They have been excluded from this analysis, though results could be material. Additionally, one carrier does not consider health status in rating. 5 This reflects moving from a 5:1 to a 3:1 age/gender slope. Using one insurer s data, for example, 85% of groups fall between -5% and 10%. As shown above, we estimate that small group market premiums will increase as a result of the ACA requirements by approximately 3% under best estimate assumptions, with a range shown for the most uncertain estimates and overall results. Employer behavior in light of the significant market changes that will take effect in 2014 creates the most uncertainty with respect to our estimates. Some of the impacts, such as age factors, have no overall impacts but the impact to a particular group can be significant. Due to the law of large numbers, the smallest groups will experience the most significant premium impacts. Each of the requirements outlined in Table 4 are discussed in detail throughout this report. July 31, 2012 Page 9

13 Small Group in Associations All of the ACA requirements expected to impact small groups will also impact small groups currently being covered through associations. Under current law association groups are permitted to use experience rating techniques. Since these rating techniques will no longer be permissible under the ACA, these groups may experience more significant impacts to premium levels. As mentioned for the traditional small group market, the smallest of these groups will experience the most drastic premium changes. Combining the Small Group Submarkets Under the ACA, the current small group and the small group association markets will merge, thereby equalizing premium levels. To understand the impact of merging these markets we compared the normalized allowed costs for the two submarkets, adjusting each for the impact of ACA requirements. Costs were further normalized to account for any differences in age, geographic concentrations and other rating variables. We estimate that merging these submarkets into one new small group market will change post-aca premiums as follows: the current small group market premiums will increase by 1% and post-aca small group association premiums will decrease by 6%. 1.3 Overview - Merger of the Individual and Small Group Markets Using the best estimate normalized post-aca allowed costs for the new individual and small group markets, premiums for the new small group market are estimated to decrease by approximately 4% and premiums for the new individual market would increase by approximately 4% if these markets were merged. Note that if the high estimate premiums are used for the new individual market then the post- ACA premiums for the new small group would decrease by approximately 1% and the premiums for the new individual market would increase by approximately 1% (assuming the two markets were merged). These estimates should be viewed as a comparison to if the markets were not merged post-aca, rather than a comparison to the current rates in the markets. These results are based on enrollment projections showing the new small group market would make up just over half of the merged individual and small group market. This impact accounts only for the differing morbidity of the markets. The pricing assumptions applied to administrative loads and the net impact of reinsurance will affect the actual results. These factors are discussed in more detail in the report. July 31, 2012 Page 10

14 1.4 Overview - Important Caveats Estimates of future premiums and programs over three years into the future under a set of changes as sweeping as the ACA are inherently uncertain. The following issues were most notable in creating this uncertainty: 1. Our analysis was completed with 2011 market information. Even in the absence of ACA changes, the market will change significantly over the course of three to six years (2011 to 2014/ 2016). 2. Important decisions have yet to be made regarding the health insurance exchange (HIX), including how active of a purchaser the state will be, oversight responsibilities, adverse selection avoidance strategies, risk adjustment methods, and others. These decisions will all affect competition among carriers, carrier rate setting methods and assumptions, and member behavior. 3. Pending guidance and regulations from the federal government may affect the appropriateness of our estimates. 4. Rates, especially in 2014 through 2016, depend on how health plans think costs will change under the ACA reforms and population expansions, not necessarily on how costs actually change. Results and information as presented in analyses such as this are important to communicate with the health insurance carriers. Feedback from these carriers on information they will find useful (e.g., state rules around rate review, information on the uninsured population, risk adjustment simulations, and others) will be critical to avoid irrational pricing. 5. Rate changes in the small group market and other financial incentives may drive employers to make unanticipated decisions around coverage. 6. The currently uninsured population will likely represent a significant portion of the individual insurance market in While migration assumptions were made, a more detailed Who Goes Where analysis should be completed based on current ACA requirements to better understand expected migration under the ACA. Shifts in enrollment may occur differently than what has been projected if the rate changes in the small group market and other financial incentives drive some employers to drop coverage. 7. Pent up demand has been shown to significantly increase costs in the first year of enrollment for those previously uninsured. Our estimates do not reflect estimates for pent up demand in 2014, since the effect is uncertain and may be offset by reduced utilization as members may not fully understand new or increased coverage. 8. Due to the limited scope of our work and timing requirements, we requested and received summary level market information from the carriers, rather than detailed data which would have allowed more validation and refined estimates. We did not audit the data supplied. 9. The behavior of individual members and employers is difficult to predict. 10. It is difficult to predict the number and impact of grandfathered plans. The more individuals and small groups that stay enrolled in grandfathered plans, the less of an impact the ACA guaranteed issue rules will have. However, the more grandfathered plans that remain, the higher the absolute level of non-grandfathered rates since grandfathered plans are assumed to have favorable risk July 31, 2012 Page 11

15 pools. Further, the impact of merging the individual and small group market could be skewed if the proportion of enrollment in grandfathered plans is very different between individual and small group. 11. Any adjustment to costs and premiums resulting from revised contracting post ACA was not considered. Reduced contract costs might result from eliminating the level of uncompensated care for uninsured residents or increases in contracting may be necessary because of provider capacity limits. 12. We did not attempt to model the impact of state mandatory benefits above and beyond the federal requirements of essential benefits, including how they will be reflected in the small group and individual markets after implementation of the ACA. 13. We have assumed that associations consisting of small employer groups will become part of the new small group market. 14. We have assumed that Oregon will move forward with Medicaid expansion despite the recent Supreme Court decision which no longer makes this expansion mandatory. 15. We have assumed that the Healthy KidsConnect members will become part of the new individual market. 16. There may be additional assessments, such as an assessment by the Health Insurance Exchange to fund the operation of the Exchange. 17. Some individuals may enroll in catastrophic plans, which have less restrictive cost sharing requirements. The impact of these plans on the estimates of the ACA changes is not expected to be significant, but would lower the premium increase estimates. July 31, 2012 Page 12

16 2. ANALYSIS OF THE OREGON INDIVIDUAL MARKET The following sections focus on the analysis of the current individual market. Comments specific to the other individual submarkets (OMIP/FMIP, portability and Healthy KidsConnect) can be found in Appendices A, B and C. At the end of section 2.4, the impact of merging the individual submarkets is discussed. Section 2.5 (Premium Tax Credits and Cost Sharing Subsidies) is based on the entire new individual market. 2.1 Summary of Individual Analysis We expect ACA provisions to produce the following current individual market changes for 2014 compared to 2011: 1. Overall increase of 13% to individual premiums due to ACA, with possible outcomes ranging from 2% to 30%. Incorporating the impact of the submarket merger, the overall premiums are expected to increase 27% to 55%. The impact to each individual will vary significantly based on the benefit plan and type of rate they currently have with an insurer, and will vary as a result of premium tax credits that the individual may be eligible to receive. 2. Qualified low income individuals may also be eligible to receive cost sharing subsidies to offset out of pocket costs beyond the premium, if they enroll in a silver plan through the exchange. 3. A compression of rates due to changes in the maximum age rate difference. The current market has age ratios greater than the 3:1 ratio permitted under the ACA. While this change will not have an overall impact on the total premiums, at the individual member level it will cause rates for younger members to increase and rates for older member to decrease. 4. The individual market will see a significant influx of new enrollees coming mostly from the current uninsured population. While the estimated impact of this change contains the most uncertainty, it drives the majority of the overall expected change in premiums. While the rating approaches for the seven largest insurers analyzed are relatively similar, currently enrolled individuals will be impacted differently based on their insurer, as well as the variables noted above. 2.2 Current Regulations and Market Composition Market Composition Seven insurers make up approximately 90% of the Oregon individual market. There are other smaller insurers in the market but our analysis focused on the seven insurers with the largest market share. July 31, 2012 Page 13

17 Table 5: Insurers and Members in the Oregon Individual Market Market Enrollment Individual Portability OMIP/FMIP 2011 Average Members Healthy KidsConnect Total % of Total FMIP % Health Net 3,914 1, ,933 3% Kaiser 10,018 4,884-1,002 15,904 8% ODS 26, ,808 14% OMIP ,697-12,697 7% PacificSource 12, ,008 16,377 8% Premera / LifeWise 23, ,225 12% Providence 10,898 1, ,480 6% Regence 55,325 5, ,208 32% United % Other * 15,915 2, ,530 10% Total 159,149 17,500 13,598 4, , % % of Total 82% 9% 7% 2% 100% * Other is based on historical percent of members not covered by the largest seven insurers. Table does not include the Uninsured which are estimated to be around 636,000. Most of these uninsured will be eligible for other coverage (e.g. Medicaid). Comparing the detail data provided by the insurers to the overall market enrollment listed on the Department s website, the analysis presented in this report represents approximately 81% of the individual market, approximately 79% of the portability market and 100% of OMIP/FMIP and HKC. Note that the results of this analysis assume the remaining market and rating characteristics are similar to the insurers included in this analysis. Any variations could significantly impact the results. According to research conducted by the Kaiser Family Foundation, in , approximately 6% of the Oregon population was insured through the individual market, while 17% were uninsured, 49% were covered under group coverage, and 27% were insured through public programs (Medicaid and Medicare). 2 Note that the Kaiser Family Foundation also states that the number of individuals insured through the individual market was approximately 236,700 in This enrollment is consistent 2 Oregon: Health Insurance Coverage of the Total Population, states ( ), U.S. (2010), July 31, 2012 Page 14

18 with the Department s 2009 estimates for individual, portability and the Oregon medical insurance pool combined. A summary of individual market benefits, commissions and loss ratios (for 2011) is provided below. 1. Benefits Benefits are separated into the benefits covered and the actuarial value (AV), which reflects the relative richness of the plan design. The AV is calculated by dividing the insurer s expected claims cost by the total covered amount of health care expenditures under ACA, including any essential benefits. Maternity is a mandated benefit in Oregon and thus, there is no impact to this benefit being an essential health benefit (EHB) under ACA. Not all insurers currently cover prescription drugs and mental health/substance abuse (MH/SA) services in the individual market. Approximately 23% of members are in plans with no drug coverage while an additional 3% are in plans with limited generic only coverage. For MH/SA, 18% of members are in plans with no MH/SA coverage, another 18% have no MH but have SA coverage, 39% have coverage but with limited hospital days and office visits. Only 18% of the market has full parity MH/SA benefits. No plans currently cover pediatric dental and vision to the level required under ACA. AVs for the current plans range from minimal coverage of 37% to as much as 92%, after accounting for the addition of essential benefits that must be covered under the ACA. 2. Broker/Sales Commissions Overall, all of the insurers included in our review are reflecting 2011 commissions on their individual book of business that are 2-6% of premium. 3. Loss Ratios 2011 actual loss ratios (paid claims over premium) for the individual market were fairly similar for five of the seven insurers. Five insurers had a loss ratio of around 68 74%, one insurer s individual book of business had a loss ratio that is already around the ACA minimum Medical Loss Ratio (MLR) of 80% and one insurer had a loss ratio greater than 80%. After adjusting for taxes and the OMIP assessment, the loss ratio range for the five insurers increased to 72 79%. Note that none of the insurers is fully credible under the ACA rules and thus, only a portion of any calculated rebate would need to be refunded. Regulations Oregon currently only allows rate variation based on age, family size and region. In general, there are no rate variations on individual premium for gender and health status. The following highlights current regulations in the Oregon market. July 31, 2012 Page 15

19 1. Underwriting medical underwriting is used in the Oregon individual market, with a denial rate of approximately 25% in the fourth quarter of This relatively high rejection rate indicates that the members accepted into the current individual market are likely to be significantly healthier than the average individual. As mentioned above, Oregon does offer a high risk pool via the OMIP/FMIP program. 2. Mandatory Benefits Oregon requires coverage for many benefits including, but not limited to, the following: clinical trials, contraceptives, hearing aids for children and dependents, HPV vaccine, mastectomy-related services, oral anticancer medications (if cancer chemotherapy is covered), services to treat pervasive developmental disorder in children, pregnancy and childbirth, non-fda approved drugs (if prescription drug coverage), medically necessary prosthetic and orthotic devices, and tobacco cessation programs. A more comprehensive list can be found on the state s website Rate Filings Rating actions are required to be filed and approved by the state prior to implementation. 4. Rating Variables two rating variables are used in the Oregon individual market (discussed below). Other rating variables are not allowed in the individual market, including, but not limited to: preexisting condition exclusions, rate-ups for substandard health and smoking status, and durational variations. a. Demographic Rating the individual rating variables are gender-neutral in Oregon. The insurers rate based on age. The current age ratio of rates varies significantly by insurer. In the individual market, the oldest to youngest adult insured ratio is between 3.3:1.0 and 5.2:1.0. b. Geographic Rating Two insurers use geography factors to rate individual premiums. The ratio of most to least costly region varies between 1.09:1.0 and 1.2:1.0. All rating variables noted above that are currently used will be allowed under ACA but these variables will have limits around the factors that may be utilized. While smoking factors are not currently utilized in the Oregon individual market, they will be allowed under ACA. Because of challenges associated with self-reported data and the cost of testing for smoking, it is not clear if issuers will employ a smoking adjustment. 3 Oregon Department of Consumer & Business Services: Health Insurance Quarterly Enrollment Reports July 31, 2012 Page 16

20 2.3 Data Received The analysis in this report is primarily based on data provided by the insurers and the state. This information includes but is not limited to: 1. Detailed benefit plan information for plans representing at least 80% of the individual, portability and Healthy KidsConnect books of business, as well as 100% of the OMIP/FMIP program. The detailed information includes: a earned premiums, allowed and paid claims, and member months by benefit plan. The same data elements were also provided in aggregate for the balance of the remaining plans in the insurer s small group book of business. b. High level cost sharing and covered services information for each benefit plan. 2. Summary of member months, premium, claims and allowed cost experience by line of business (small group, individual, portability, Healthy KidsConnect, OMIP/FMIP, size groups) and product type. 3. Rating factors by age band, gender, family size/status and line of business. 4. Member months by gender and age band. 5. Administrative loads, by admin component, as a percent of premiums. While most of the data received were for plans that are open, (currently accepting new enrollment) approximately 22% of the individual market is in a grandfathered plan. Only three insurers have grandfathered individual plans with over 70% of each insurer s members currently in grandfathered plans. For the insurers with grandfathered plans, two of the three have higher loss ratios in their grandfathered plans while the other has lower. The actuarial values are similar within the insurers, with one insurer s grandfathered plans having slightly less rich cost sharing than its non-grandfathered counterparts. Combining the data for the three insurers, loss ratios and actuarial values are both lower in the grandfathered plans than non-grandfathered plans. For the purpose of this report we have assumed all grandfathered plans will be impacted by the ACA. That is, it is assumed these plans will make all necessary changes to be ACA compliant, whether or not they are grandfathered. This assumption is conservative and any variance from this assumption will lessen the number of individuals who are impacted by the ACA requirements. To the extent insurers offer and sell new plans in that incorporate some or all aspects of ACA, this would also lessen the number of members impacted as well as the magnitude of the impact. July 31, 2012 Page 17

21 2.4 ACA Impact on Individual Market The ACA includes a provision for states to develop an exchange through which individuals can purchase health insurance. The ACA also includes significant new underwriting and rating requirements for the individual market. The purpose of this section is to analyze the estimated impact of these new requirements on current individual premiums. Since the ACA dictates that rates need to be the same for insurers who offer insurance both in and out of the exchange, the focus of the analysis is on premiums in the overall individual market, in or out of the exchange. The following discusses the premium impact of the ACA requirements on the current individual market. Separate premium impacts for the portability, OMIP/FMIP and Healthy KidsConnect markets can be found in Appendices A, B and C. Overall current individual premiums are estimated to increase by 38%, 13% due to the ACA s underwriting and rating requirements and an additional 22% due to the submarket merger. This change is not uniform over all individuals, as some enrollees could see their premiums decrease slightly while others could see their premiums more than double before taking into consideration premium tax credits. The table below shows the overall projected change for each of the requirements. As can be seen in the table, the most significant requirement to the Oregon individual market, as measured by overall impact to premium, is the morbidity change (driven by new entrants in the market). While the age slope limitation has no overall impact, its impact to individual enrollees may be significant. The overall premium impacts (gross of premium tax credits) by ACA requirements are shown in the following sections. July 31, 2012 Page 18

22 Table 6: Changes to Current Individual Market Premiums under ACA (2014 compared to 2011) Requirement Description Best Estimate Impact ($ PMPM) 2011 Average Individual Monthly Premium $202 Best Estimate Impact (%) Essential Benefits Requirement $12 6 % Bronze Minimum Act. Value (includes Max OOP limit) $5 2% Minimum Loss Ratio = 80% -$1-1% Age Slope Limited to 3:1 $0 0% Morbidity Changes (due to new insureds) $30 15% Provider Fee $2 1% Reinsurance Program $-17-8% Elimination of OMIP Assessment $-2-1% Subtotal New Average Premium (2011 Dollars) $230 13% Submarket Merger $49 22% New Average Premium (2011 Dollars) $279 38% The above table does not include the impact of pent up demand or other rating requirements. See the Additional Requirements and Considerations section for information on these topics. Also note that due to rounding the individual components in the above table will not add to the total. Figure 1 below shows the various estimated premium changes and the corresponding percent of members affected. Enrollees receiving the greatest premium increases are likely to be younger individuals, who are currently in catastrophic benefit plans. Conversely, enrollees receiving the largest premium decreases are likely older enrollees. The premium changes included in this section reflect the average premium change from the perspective of the insurer. They do not take into account the impact of the premium tax credits that will be available to lower income individuals. The impact of premium tax credits and cost sharing subsidies is discussed later after all ACA impacts and member migrations have been considered. July 31, 2012 Page 19

23 Figure 1: Estimated Premium Changes and Percent of Members Impacted Premium Impact Essential Benefits The ACA requires that all benefit plans cover services for essential health benefits, some of which are often excluded in the current individual market. Based on the services enumerated in the ACA, the essential health benefits will include, but are not limited to, the following categories: 1. Ambulatory patient services 2. Emergency services 3. Hospitalization 4. Maternity and newborn care 5. Mental health and substance use disorder services, including behavioral health treatment 6. Prescription drugs 7. Rehabilitative and habilitative services and devices 8. Laboratory services 9. Preventive and wellness services and chronic disease management, and 10. Pediatric services, including oral and vision care Currently, all individual insurers in Oregon cover maternity. Approximately 82% of the current individually insured market is in a plan with at least some mental health and substance abuse coverage. 77% of members are enrolled in a benefit plan with prescription drug coverage. The impact of adding mandatory coverage, for those services defined above, as part of the essential benefits is estimated to be 3%. The primary drivers of the essential benefits impact are adding or increasing prescription drug July 31, 2012 Page 20

24 coverage and MH/SA to the plans that currently do not cover it to the extent required and adding pediatric vision and oral services. The required pediatric oral benefits are comprehensive, covering most services with the exception of non-medically necessary orthodontics. A high level estimate for the impact of adding habilitative services has also been included but this benefit still remains undefined. The figure below shows the range of impact by members for all essential benefits. The overall impact of adding these essential benefits to the current benefit plans is approximately 6%. Plans that currently do not have prescription drug coverage are usually plans with high cost sharing and the least covered benefits, and the combined impact of adding all essential benefits increases these plans premiums by as much as 20%. Figure 2: Impact of Essential Benefits Requirement Maximum Out of Pocket Limits Starting in 2014, the maximum out-of-pocket (MOOP) amount cannot be greater than the HSA limit ($5,950 for individual policies and $11,900 for family policies in 2011, indexed annually). The percent of individually insured members in current plans that will be affected by this requirement is roughly 65%. The impact of this change has not been explicitly determined but is included in the Bronze Requirement impact (discussed below). Bronze Requirement Beginning in 2014 there will be four primary levels of plan designs that may be offered, varying by their actuarial value. The four plans are: Bronze at 60% AV, Silver at 70%, Gold at 80% and Platinum at 90%. While not all of the specifics of this requirement have been finalized, all insurers participating in both July 31, 2012 Page 21

25 the individual market and the exchange will be required to offer one plan at the Silver level and one at the Gold level. In addition, most states will mirror these requirements in the market outside of the exchange (assuming such a market is allowed by the state). The table below shows the current distribution of plans and members by various AV levels. These AVs have been adjusted to include any essential benefits that may not currently be covered. Table 7: Current Member Distribution by AV Individual Actuarial Value Range % of Members <45.0% 0% 45.0%-55.9% 14% 55.0%-64.9% 27% 65.0%-74.9% 35% 75.0%-84.9% 20% 85.0%-94.9% 3% 95.0% or higher 0% Total 100% Given that the individual market tends to offer less rich benefit plans than the group market, it is likely that most insurers will also offer the least rich benefit plan allowable, or the Bronze level, in addition to the required Silver and Gold plans. We have assumed Bronze is the minimum benefit plan members could purchase through the exchange. For our analysis it is assumed all benefit plans would at a minimum need to meet the 60% Bronze AV level. Note that while catastrophic plans will allow some members to enroll in plans with a lower AV, it is difficult to know what portion of members will enroll in these plans. The impacts shown assume no catastrophic plan enrollment and are therefore conservative estimates. Figure 3 below shows the various impacts for members enrolled in plans that currently do not meet the Bronze requirement. July 31, 2012 Page 22

26 Figure 3 Impact of Bronze and MOOP Requirement Medical Loss Ratio (MLR) Requirements Effective January 1, 2011 insurers are required to maintain a minimum loss ratio of 80% (for the individual market) or the insurer must pay rebates back to the enrollees. It is expected that loss ratios under the ACA will be higher than traditional MLR calculations (incurred claims divided by earned premiums) given the allowable adjustments under the ACA. The calculated MLR will also be subject to credibility adjustments. Credibility adjustments will be made to account for random statistical fluctuations that are inherent when an insurer has a smaller member base. Statistical fluctuations are also possible when an insurer has higher deductible plans since more of the insurers liability resides with catastrophic claims. By 2014, for MLRs in plan years 2011 to 2013, the number of members in the calculation will be the sum of the average members in each of the previous three years. Credibility adjustments will apply to insurers that have over 1,000 but less than 75,000 members. If an insurer has less than 1,000 average members, the insurer is essentially exempt from any rebate payments. If the insurer has 75,000 or more average members, their experience is deemed to be fully credible and thus the calculated MLR will determine what, if any, rebate must be paid to their members. For plans that are partially credible (members over 1,000 but less than 75,000), an additive adjustment to their calculated MLR will be made based on the number of members. An additional adjustment may be made if the average deductible in their block of business is $2500 or greater. These adjustments will increase their calculated loss ratio, lessening the likelihood that they will owe a rebate or if they do, decreasing the amount of rebate owed. July 31, 2012 Page 23

27 To assess the impact of the MLR requirement, loss ratios for the current book of business were reviewed. The incurred claims we received were not reported using the NAIC guidance for reporting medical costs in As a result, we expect that the loss ratios will increase for the disease management, quality and fraud and abuse expenses that will be allowed as medical costs. As a check of reasonability, the loss ratios provided were compared to the non-claim expense loads documented in the rate filings, if available loss ratios were calculated by insurer, adjusting for estimated taxes, OMIP assessments and a high level estimate of additional adjustments. The loss ratios were also adjusted for credibility and deductible levels as outlined above. The change in premium was then calculated to determine what decrease in administrative expenses, if any, was needed to comply with the minimum 80% loss ratio. Based on the data of the seven insurers, the overall average impact is expected to be a 1% decrease to premiums in the individual market. After all adjustments, only two of the seven insurers were under the required loss ratio and by less than 4%. Thus, we estimate that members may see between 0% to 4% rate decreases as a result of the MLR requirements. Individual market premiums may also experience premium decreases due to administrative expenses that will likely be lower under ACA due to a decrease in insurer expenses such as underwriting and commissions. The impact of this potential impact has not been incorporated into the analysis. Demographic Factors Two significant changes will be required in 2014 for any demographic adjustments to rates. Under the ACA, rates can not differ based on gender and the maximum ratio of the highest to lowest adult rate is 3 to 1. Since gender rating is not currently utilized in Oregon, the gender-neutral requirement will not have an impact on premiums. The 3:1 age slope constraint will, however, affect premium rates in the Oregon individual market. While it is unknown what age factors the insurers will implement in 2014, and each insurer will likely develop their own factors that meet the requirement 5, an assumption was made for this analysis after reviewing current demographic factors and incorporating the new requirements. Since child rates do not fall under the 3:1 requirement, no explicit requirement was applied to child rating factors. Currently, individual insurers have adult rate ratios that vary from 3.3:1.0 and 5.2:1.0 (depending on the insurer). The overall impact of the demographic changes is expected to be premium neutral; however, the member impact is significant. The largest premium increase due to age slope changes is an 81% increase while the largest decrease is 25%. The younger enrollees will see large premium increases and 5 We expect that final rating regulations may allow states to prescribe specific factors, although doing so may limit innovation and discourage health plans from participating. July 31, 2012 Page 24

28 the older enrollees will see premium decreases. The proportion of members impacted and the corresponding impact of the age changes are in Figure 4 below. Figure 4 Impact of Age Ratio Changes Underwriting Factors Underwriting factors are designed to allow insurers to vary premiums such that they would be more in line with an individual s expected costs. This enables insurers to accept more applicants by accepting higher risk individuals at a higher premium level. Under ACA, underwriting factors or classes will no longer be allowed. Since the Oregon individual market does not rate with underwriting factors, this requirement will not impact rates. Under the ACA, tobacco factors will be allowed with a maximum premium ratio of 1.5 to 1.0 (that is, health insurers may charge tobacco users as much as 50% more than non-tobacco users for a given health plan). Since no individual insurers and only one small group insurer currently use tobacco rating, it is assumed that this rating variable will not be used in the post-aca individual market. Durational Factors Durational factors, which capture the impact of underwriting decreasing or wearing off as an individual s enrollment span continues, are not used in the Oregon individual market. Since such factors are not currently used, the ACA requirement prohibiting durational factors will not have an impact in the Oregon market. July 31, 2012 Page 25

29 Elimination of OMIP Assessment In the current market, insurers pay an assessment for the purposes of funding the OMIP/FMIP program. Assuming OMIP/FMIP are folded into the new individual market under ACA, it is assumed this assessment will be eliminated. On average, it is estimated that the elimination of this assessment will decrease current individual premiums by approximately 1%. Morbidity Changes The morbidity change in the individual market will be driven by the large number of new entrants in the individual market. Some of these new currently uninsured members will enroll due the guarantee issue provision which does not allow insurers to deny a member based on their health status. To ensure not just those in need of health care enroll, the ACA requires all individuals obtain minimum essential coverage or face a tax penalty. 6 This individual mandate to maintain health coverage will help mitigate increases to the overall risk level of the individual market (assuming the penalties are sufficient to influence individuals behavior). For individuals below 400% of FPL, subsidies will be available to cover a portion of the premium if premiums under the second lowest silver plan for that individual are greater than a particular threshold set as a function of income. The individual market will see migration not only from the uninsured but also from the small group market where members may also enter the exchange, without employee or employer penalty, if they are eligible for premium tax credits. In order to determine the impact of the incoming population we analyzed a base set of data from Current Population Survey (CPS) for a benchmark state and corresponding risk adjustment factors for a similar population. The CPS information provides insight into self-reported health status. It was seen that the uninsured tend to categorize themselves as being in worse health than the population already in the individual market. The benchmark state data was used to quantify differences in health status. Given the high denial rate (25%) in the current individual market, it is expected that current market premiums reflect a relatively healthy population and the overall premiums will need to be increased in order to cover the higher risk of both the uninsured and migrating group members. While a detailed migration analysis is needed to understand the anticipated movement of the entire insurance market, some high level assumptions were made to provide a general understanding of the potential impact of member migration into the individual market. Our high level assumptions include the following: 6 People may apply for an exemption from the individual mandate based on lack of affordable health insurance (i.e., premiums are more than 8.0% of a person s modified adjusted gross income), religious beliefs, or personal hardship. July 31, 2012 Page 26

30 1. Migration will only come from the uninsured and the small group market. No large group or Medicaid migration is assumed. 2. The existing members in the individual market will remain except for a subset of members who will be eligible for Medicaid based on CPS data. Note that if there is migration from the current group market, these members are likely to be less healthy than those in the current individual market since the small group market is already guaranteed issue. The resulting migration in and out of the individual market is summarized in the table below. The table represents best estimates of 2014 and 2016 migration and does not include the impact of the merger of the various individual submarkets (e.g. OMIP/FMIP). Wakely used varying assumptions by year and under different scenarios to develop our ranges of impacts. Table 8: Expected Individual Member Migration Individual Migration 2014 Best Estimate 2016 Best Estimate Current Market 159, ,149 Medicaid Eligible -17,266-17,266 Small Group 15,000 20,000 Uninsured 58, ,021 Total 215, , Nationally, subsidy-eligible individuals, who constitute the majority of the uninsured, tend to have a higher risk profile than non-subsidy-eligible individuals. As a result, the more subsidy-eligible individuals that migrate to the individual market, the more likely premiums will increase. For Oregon, these uninsured individuals reported health that is slightly less favorable than the national average. The Oregon subsidy-eligible, uninsured individuals consistently reported a similar or less favorable health status than their non-subsidy-eligible counterparts. The Oregon individuallyinsured, subsidy-eligible reported a similar or more favorable health than their non-subsidy-eligible counterparts. See the table below for a comparison of self-reported health status for Oregon and nationally by income level. Note also that across all income levels and insured status, Oregon self-reports slightly less healthy than the national average. Even though the Oregon data is similar to the national data, a blend of national and Oregon data were used to develop morbidity factors for subsidy and nonsubsidy eligible individuals in both the insured and uninsured markets. July 31, 2012 Page 27

31 Table 9: Self-Reported Health Status Comparison Insured Status Self-Reported Health Status National Subsidy Eligible (FPL < 400%) Non-Subsidy Eligible (FPL > 400%) Oregon Subsidy Eligible (FPL < 400%) Non-Subsidy Eligible (FPL > 400%) Insured-Individual Good, Very Good or Excellent 94% 97% 97% 93% Insured-Individual Fair or Poor 6% 3% 3% 7% Uninsured Good, Very Good or Excellent 88% 93% 84% 91% Uninsured Fair or Poor 12% 7% 16% 9% Combined Good, Very Good or Excellent 89% 95% 86% 92% Combined Fair or Poor 11% 5% 14% 8% Ultimately, an aggregate 15% increase in rates is expected due to the influx of new individuals. This impact varies by year and based on the migration and morbidity assumptions used. The overall impact is expected to be 10% to 30% for all years, with a range of 10% to 25% and a best estimate of 15% in This assumption is not only the most significant, but also contains the most uncertainty. As a result, it is important to fully understand the implications of this assumption. In 2014, migration to the individual market is expected to be less than the ultimate migration due to smaller penalties and the newness of the reform. In 2014, we expect that 5-15% of currently uninsured residents will enroll in the individual market. In 2016, this increases to an expected 10-25% of the currently uninsured. While fewer uninsured members are expected to migrate in 2014, those uninsured that do migrate are likely to be less healthy and have a higher morbidity than those who choose not to enroll. Thus, lower migration numbers are offset by higher morbidity factors. As a result, the overall expected morbidity impact, taking into account the number and health status of the new enrollees, increases only slightly over the three years. Tables 10 and 11 below show the expected impact of morbidity changes due to the new entrants into the individual market. Table 10 shows two scenarios assuming a lower migration assumption, which is more likely in the first year or two. The two scenarios use different morbidity factor assumptions since the actual health status of the new enrollees is largely unknown. Table 11 also provides two different morbidity factor scenarios but assumes a higher uninsured migration, which is more likely in Note that the morbidity factors for the uninsured are lower in the high migration table. This is due to the assumption that the less healthy are more likely to enroll in the individual market in the initial years compared to their healthier counterparts who may be willing to pay the penalty rather than the cost of insurance. July 31, 2012 Page 28

32 Table 10: Morbidity Changes with Low Migration Assumption Current status Subsidy-eligible 2010 Average Members Low Morbidit y Assumpt ions % Migrat ion Fut ure Ind Market Enrollment Morbidit y Fact or High Morbidit y Assumpt ions % Migrat ion Fut ure Ind Market Enrollment Morbidit y Fact or Uninsured Subsidy-eligible 567,850 10% 56, % 56, Uninsured Not subsidy-eligible 68,150 5% 3, % 3, ESI Subsidy-eligible 816,335 1% 10, % 10, ESI Not subsidy-eligible 1,050,365 0% % Individual Subsidy-eligible 100,832 83% 83, % 83, Individual Not subsidy-eligible 58, % 58, % 58, Morbidity change to current insured individual market 212,200 10% 212,200 27% Table 11: Morbidity Changes with High Migration Assumptions Current status Subsidy-eligible 2010 Average Members Low Morbidit y Assumpt ions % Migrat ion Fut ure Ind Market Enrollment Morbidit y Fact or High Morbidit y Assumpt ions % Migrat ion Fut ure Ind Market Enrollment Morbidit y Fact or Uninsured Subsidy-eligible 567,850 25% 141, % 141, Uninsured Not subsidy-eligible 68,150 15% 10, % 10, ESI Subsidy-eligible 816,335 2% 20, % 20, ESI Not subsidy-eligible 1,050,365 0% % Individual Subsidy-eligible 100,832 83% 83, % 83, Individual Not subsidy-eligible 58, % 58, % 58, Morbidity change to current insured individual market 314,193 10% 314,193 26% Pre-existing Condition Limitations As noted above, Oregon does not allow insurers to exclude coverage for certain conditions that are preexisting at the time a person applies for insurance (called pre-existing condition limitations). Since the Individual market in Oregon does not use pre-existing limitations, the ACA requirement prohibiting this exclusion will not impact premiums. The Health Insurance Provider Fee The insurer fee applies to for profit health insurance companies commercial and public program enrollees. The insurer fee does not apply to third party administrators. The insurer fees to be collected by year according to the ACA are as follows: July 31, 2012 Page 29

33 Table 12: Insurer Fee National Contribution Requirements Year Statutory Insurer Fee Amount (in $B) 2014 $ $ $ $ $ $ $ $ $ $22.1 The insurer fee will be assessed equally as a percentage of applicable premiums (besides excluding not for profit companies, the first $25 million of premium, and 50% of premium from $25 to $50 million for an organization is not assessed). The insurer fee is not a tax deductible expense for federal corporate tax purposes. Wakely developed estimates of the insurer fee and its impact on premiums. Because the average impact of the insurer fee on premium rates depends on the proportion of not for profit and tax exempt premiums, and Oregon has a high proportion of not for profits, the impact of the insurer fee on premium rates in Oregon is less than the national average. The following table shows the best estimate impact of adjusting the national average values to Oregon: Table 13: Wakely Estimates of Oregon Premium Impact by Year Insurer Type Distribution of Members Provider Fee For Profit 41.3% 1.96% 2.53% 2.34% Not for Profit 58.7% 0.00% 0.00% 0.00% Total 100.0% 0.81% 1.05% 0.97% The following table shows the results of our national estimates: July 31, 2012 Page 30

34 Table 14: Wakely Estimates of National Insurer Fee by Year Year Insurer Fee as % of Total Premiums Insurer Fee as % of Premiums Subject to Fee % 1.96% % 2.53% % 2.34% % 2.74% % 2.67% % 2.53% % 2.39% % 2.62% % 2.86% % 3.13% The first column of estimates (Insurer Fee as % of Total Premiums) represents an estimate of the increase in average premiums by year due to the insurer fee. The second column of estimates (Insurer Fee as % of Premiums Subject to Fee) represents the assessment insurers will pay on premium that is subject to the fee (premiums subject to the assessment are reduced for tax exempt and smaller insurers, and eliminated for non-profit insurers). The estimates in the first column of the table above for the insurer fee are lower than the estimates developed by Oliver Wyman. 7 Our estimates are very similar to estimates developed by Milliman. 8 We believe that the difference between our estimates and the Oliver Wyman estimates is likely a different assumption regarding the portion of total members who are covered under self-funded plans, since that is the assumption most likely to produce such a large difference. The model includes baseline estimates of enrollment, per capita premium and total premium by state for the following six key business segments using the Urban Institutes HIPSM: 7 Carlson, Oliver Wyman, October 31, 2011, Estimated Premium Impacts of Annual Fees Assessed on Health Insurance Plans 8 Meerschaert & Doucet, January 31, 2012, PPACA Health Insurer Fee Estimated Impact on State Medicaid Programs and Medicaid Health Plans, July 31, 2012 Page 31

35 Table 15: Application of Fees by Market Segment Market Segment Insurer Fee (Premium) Individual Small Group Large Group - Fully Insured Medicaid Managed Care Medicare Advantage Large Group - Self Funded X X X X X As shown above, the insurer fee applies to premiums in all of the above market segments except selffunded large group. For the insurer fee, an adjustment is made for tax exempt and non-profit insurers and for the exclusion of the first $25 million in premium and 50% of premium between $25 million and $50 million. The insurer fee impact as a percentage of premiums subject to the fee is calculated as: Only half of the premium from tax exempt organizations is subject to the insurer fee. Not for profits are fully exempt from the insurer fee. Wakely has estimated that approximately 16% of premiums are from tax exempt organizations and 12% are from not for profits. The higher the proportion of tax exempt or non-profit premiums, the higher the percentage assessment on other organizations (the premium base is reduced). The ACA insurer fee includes relief for smaller insurance companies by exempting the first $25 million in premiums and only assessing 50% of premiums between $25 million and $50 million. This effectively increases the assessment percentage for amounts over $50 million and therefore for larger insurance companies. We used the NAIC supplemental healthcare exhibit for 2011 to estimate the impact of this exemption. The assessment percentage is approximately 1% higher on average because of this exemption. A number of key assumptions drive the results, roughly ordered in terms of materiality: 1. Actual shifts across the markets, especially to or from the self-funded employer group market. Any member shifts between for and not for profit companies will also impact the overall impact. July 31, 2012 Page 32

36 2. Impact of the ACA, and state specific market / HIX structural decisions on enrollment take up in the various markets. 3. Proportion of Medicaid expansion that enrolls into managed care. 4. Impact of the ACA on medical costs and therefore premium rates. 5. Underlying medical cost and utilization trends by market. Reinsurance The reinsurance program under the ACA is a temporary program that will operate from 2014 through The reinsurance program is intended to protect health plans operating in the individual market from specific high cost individuals. States that establish an exchange, such as Oregon, may operate the reinsurance program or they may allow HHS to operate the program. The ACA includes the following nationwide requirements for reinsurance contributions: 2014 = $10 billion 2015 = $6 billion 2016 = $4 billion In addition, required national contributions to the U.S. Treasury to provide health reform funding are as follows: 2014 = $2 billion 2015 = $2 billion 2016 = $1 billion Preliminary modeling suggests that the assessment on issuers will be approximately 1% of 2014 premium (or medical cost for self-insured) for the reinsurance-only portion of this assessment. This assumes that the assessment will be set based on a national calculation, which would increase by 20% in 2014 if the Treasury contribution is included. The 2014 reinsurance assessment should decrease individual market premium rates in Oregon by approximately 7% to 9%, changing to a decrease of 2% to 3% by These ranges of estimates depend on a number of factors, including: the size of the individual market, and actual morbidity level in the individual market. July 31, 2012 Page 33

37 The following table shows preliminary impact estimates of the reinsurance assessment on individual market premiums in Oregon and the US. Table 16: Estimates of Impact of Reinsurance to Premiums Description Net Assessment (Reinsurance Only - Not Treasury Contribution) Net Impact to Individual Market Premiums (US) Net Impact to Individual Market Premiums (Oregon) Higher Estimate of Individual Market Lower Estimate of Individual Market % 0.7% 0.4% 1.3% 0.7% 0.4% -8.9% -4.4% -2.6% -13.7% -6.6% -3.6% -7.3% -3.6% -2.1% -9.3% -4.6% -2.6% Combining the Individual Submarkets In addition to the large membership increase the individual market will experience due to the influx of uninsured and small group members, other current pools of members will likely fold into the individual market. The merging of the current individual market with the portability, OMIP/FMIP and Healthy KidsConnect markets will create an even larger individual market. To understand the impact of merging these markets we compared the normalized allowed costs for the various submarkets, all adjusted for the impact of ACA requirements including the migration from the uninsured into the current individual market. ACA impacts on each of the other individual submarkets (portability, OMIP/FMIP and Healthy KidsConnect) can be found in Appendices A, B and C. Costs were further normalized to account for any differences in age, geographic concentrations and other rating variables. The table below shows the estimated 2014 impact of merging these markets. July 31, 2012 Page 34

38 Table 17: 2014 Estimated Premium Impact of Merging the Individual Submarkets Individual Submarket Individual (Includes New Enrollees) Post ACA Allowed PMPM Projected Members Impact of Sub- Market Merger Premium Impact due to ACA Requirements Total Premium Impact $ ,812 22% 13% 38% Portability $665 17,500-39% 50% -8% OMIP/FMIP $1,234 13,598-67% 141% -21% Healthy Kids $391 4,010 4% -9% -5% Combined $ ,920 These submarket merger impacts reflect the premium impact due to morbidity differences. Since OMIP/FMIP and portability premiums are currently subsidized, the premium impact is related to the full premiums that would need to be charged to cover expenses rather than the impact to the currently subsidized premiums is included in the Impact due to ACA Requirements. Combining the merging of the individual submarkets with the impact of ACA requirements (discussed above and in the appendices) results in the following post-aca premium changes: the current individual market will increase by 38%, post-aca portability premiums will decrease by 8%, post-aca OMIP/FMIP premiums will decrease by 21% and Healthy KidsConnect premiums will decrease by 5%. 2.5 Premium Tax Credits and Cost Sharing Subsidies The ACA provides for premium tax credits and cost-sharing subsidies for eligible individuals in the exchange who have incomes less than 400% FPL. For these individuals, premiums are limited to a sliding scale of 2% for those at 100% FPL to 9.5% of income for those at 400% FPL. Cost-sharing subsidies limit the members cost-sharing and in effect increases the AV of the selected benefit plan without increasing the premium amount. These subsidies for low income individuals can be significant. This section describes the expected impact to both the premium and cost sharing from the individual s perspective. The analysis considers only individuals earning above 133% FPL, 9 since individuals below this level will be eligible for Medicaid and will therefore not be eligible to receive premium tax credits or cost-sharing subsidies within the exchange. 9 Legal residents who have not resided legally in the United States for five years will continue to be ineligible for Medicaid/CHIP, but may be eligible for subsidized coverage through the Exchange. As a result, those legal residents with income below 138% will also be eligible for coverage through the Exchange. July 31, 2012 Page 35

39 If a state chooses to establish a Basic Health Plan (BHP), those with incomes under 200% FPL will be required to enroll in the BHP for their subsidies instead of accessing them through the exchange. Since the state of Oregon does not currently plan to establish a BHP, this enrollment requirement will not apply. If the state chooses to implement a BHP then additional analyses will need to be completed on the relative risk and size of the BHP eligible population. Premium Tax Credits Premium tax credits will be available on a sliding scale based on the income level of the individual. Qualifying individuals at 133% of FPL will receive credits so that their premium costs are not above 3% of their income. Credits will reduce as income levels increase to the point that individuals at 400% of FPL are not paying more than 9.5% of their income for health insurance premiums. There are no premium tax credits available above 400% FPL. The following table show the limits outlined in the ACA for a family of one. Since FPL levels vary by family size, the limits will also vary by family size. Given these constraints, for any given FPL, younger individuals are less like to qualify for premium tax credits (or qualify for less) as compared to their older counterparts. Furthermore, single individuals qualify less than families. Table 18: Premium Tax Credit Levels by Income Level Income Level (% of FPL) Max amt of income for premiums Max Monthly Premium Contribution* Max Annual Premium Contribution* 133% 3.0% $36 $ % 4.0% $54 $ % 6.3% $114 $1, % 8.1% $184 $2, % 9.5% $259 $3, % 9.5% $302 $3, % 9.5% $345 $4, %+ N/A $0 $0 *as of 2011 The premium tax credit amount will be based on the second lowest cost silver plan available to the individual in an exchange; although the individual is able to enroll in other than silver level plans and still receive the tax credit. Individuals enrolling in more expensive plans than the second lowest silver plan will have to pay additional premium amounts out of pocket. On average, the expected impact of the premium tax credits is an effective subsidy of approximately 30% of premiums for the new individual market (including uninsured) in Should the actual population enrolling be significantly different than what was modeled, the impact may be significantly impacted. The following table shows the expected impact of premium tax credits by income range. We July 31, 2012 Page 36

40 utilized CPS data to determine the current population mix by age and FPL and plan specific data to determine splits by coverage tier (e.g. sub only, family, etc.) and average family size. This information was then used to determine the average premium tax credit. The impact of the subsidy, shown in the table below, is the average for the population in each income range. The credits are calculated based on the average post-aca premium, as compared to the second lowest silver. If an individual enrolls in a more expensive plan, the tax credit as a percent of premium will be reduced. We also separate the population into those that are currently insured and those that are expected to become newly enrolled. There are more individuals in the higher income ranges in the current population, so their premium tax credit is expected to be lower than that of the incoming population. The premium tax credits displayed in the table below should be considered as the best estimate average subsidies. The figures in the table are calculated based on the weighted average expected premium across all ages and coverage tier compared to the maximum premium for the income range. For a given income level, younger people who are more likely to have lower premiums will be more likely to have a lower premium tax credit, if any. Conversely, older people having higher premiums will be more likely to have greater premium tax credits. We are reflecting credits for a representative distribution of family sizes, based on splits provided by the insurers. FPL Table 19: Effective Premium Tax Credit by Income Level Individual Premium Impacts # of Members % Eligible for Tax Credits Before Premium Tax Credits After Premium Tax Credits Currently Insured Newly Enrolled Currently Insured Newly Enrolled % 23% -50% 12,639 8, % 100% % 24% -37% 29,181 18, % 100% % 27% -16% 30,079 19,424 91% 95% % 25% 0% 17,524 11,316 67% 53% % 22% 8% 8,741 5,645 53% 40% % 34% 29% 6,080 3,927 51% 36% 400%+ 20% 20% 72,747 6,613 0% 0% Total 24% -10% 176,991 73,929 50% 74% The overall premium change without premium tax credits (24% in Table 19) is the weighted average impact of the ACA requirements on the merged individual submarkets (i.e. the new individual market). The actual impact to each of the submarkets and members within these submarkets will vary. July 31, 2012 Page 37

41 Cost Sharing Subsidies Individuals who qualify for premium credits and are enrolled in a silver plan in the exchange will also be eligible for assistance in paying their cost sharing. Any plan in the exchange will already have a limit on the MOOP such that it cannot exceed the high deductible health plan limit ($5,950 in 2011). The cost sharing subsidies will further reduce these MOOP limits by two-thirds for individuals up to 200% of FPL, by one-half for individuals between 200% and 300% of FPL, and by one-third for individuals between 300% and 400% of FPL. Other cost sharing such as deductibles, coinsurance, and copays will be further subsidized, if necessary, to ensure that the health plan and subsidies cover the percentages of allowed health care expenses as shown in the following table. The individuals would be responsible for the remaining amount of allowed expenses reflected in the final column of the table below. Table 20: Cost Sharing Subsidy by Income Level Income Level (% of FPL) % of Allowed Expenses Covered by Plan and Subsidy % of Allowed Expenses Covered by Individual % 94% 6% % 87% 13% % 73% 27% % 70% 30% % 70% 30% % 70% 30% 400%+ 70% 30% The silver plan level contemplates an actuarial value of 70%, consistent with the covered expenses for 251% FPL and above. Our assumption is that only individuals with incomes of 250% of FPL or less will receive the cost sharing subsidy. 10 The following table shows the subsidy as a percent of expected cost sharing for each income level. Given the expected population mix, the overall impact is expected to be a 23% reduction in the cost sharing burden on individuals. The currently insured population, due to a higher average income level, is expected to have a lower subsidy than the incoming population July 31, 2012 Page 38

42 Table 21: Effective Cost Sharing Subsidy by Income Range Income Range Cost Sharing Subsidy % 85% % 54% % 9% % 0% % 0% % 0% 400%+ 0% Avg. Cost Sharing Subsidy 23% 2.6 Additional Requirements and Considerations The previous sections (2.4 and 2.5) outlined our analyses with respect to requirements that had a quantifiable and significant impact on individual premiums. There are additional requirements and considerations that have already been implemented or may impact premiums and rating practices and are worth noting. September 23, 2010 Requirements A few regulations that impacted premiums took effect September 23, 2010 (i.e., six months after enactment of the ACA). Identifying the impact of these requirements is not part of our analysis as the focus was on the impact of implementing all of the ACA requirements yet to be incorporated. We did, however, consider that the impact of these items may already be reflected in the data and filings received. The following September 2010 regulations likely had some marginal impact on premiums: 1. Preexisting condition exclusions for children under age 19 is no longer allowed % coverage for specified preventive benefits. 3. Lifetime benefit limits for essential health benefits is no longer allowed. 4. Annual limits on essential health benefits are restricted. Additional rating restrictions for the individual market Rating area ACA requires that area rating will be determined by each state, contingent on review and approval by the Secretary of HHS. Since it appears that area factors are not currently utilized and since the state is small in size, this is not expected to affect current rating practices. July 31, 2012 Page 39

43 Family structure The current proposed rules recommend a four tier rating structure. Since this will be consistent across the market, this is not expected to have a premium impact. Grandfathered plans These plans are exempt from many of the ACA requirements and thus will dilute the impact of the ACA requirements on the individual market. Roughly 22% of the current individual members are in grandfathered plans. However, it is unknown how many of these plans will remain grandfathered. Pent Up Demand There will be new enrollees in the individual market that were previously uninsured. Prior to enrolling in coverage, these people would have been paying out of pocket for any medical costs, and it can be assumed that they did not treat minor health problems and did not receive preventive care due to cost. Once they are covered in the Exchange, there may be an increase in utilization for this population as they will be more likely to afford to have minor issues treated or utilize preventive care visits. Based on analysis of programs serving a similar population, it appears that the claims are higher in the first six to 12 months that an individual is newly insured. After this initial period, the claims experience tends to revert to the average level for the market. Within the first year, claims for those newly insured may be 3-4% higher than average. We have not factored in any increase in the premium to account for pent up demand. In the current block, insurers already are reflecting the impact of pent up demand to a degree as they already cover individuals that may be previously uninsured. The population of individuals that will be covered by the exchange is expected to consist of people coming from an insured environment (either through individual or group insurance), individuals from the high risk pool, and individuals that are currently uninsured. Provider Reimbursement The ACA is expected to significantly expand coverage both by reducing out of pocket costs and by allowing or requiring individuals who previously didn t have insurance coverage to obtain coverage. When individuals lack coverage or have high out of pocket costs, hospitals, physicians and other healthcare providers sometimes provide care without being fully compensated. Because the proportion of individuals without coverage is expected to decrease and individuals will generally have lower out of pocket costs under the ACA reforms, the proportion of uncompensated care delivered by providers should decrease. This decrease in uncompensated care could be expected to allow providers to reduce their reimbursement rates and still receive the same level of total revenue. However, the influx of newly insured members who may now seek out care more than they had prior to having coverage will increase demand for healthcare services and therefore providers. This increased demand may offset some of the effect, allowing providers to increase the reimbursement rates they receive from insurance companies. The uncertainty associated with these complex market changes is too large for Wakely to July 31, 2012 Page 40

44 provide credible estimates. However, various sources indicate that payments to Oregon providers could increase by hundreds of millions of dollars per year over and above any underlying healthcare cost trends. Disproportionate Share Hospital (DSH) payments will decrease significantly starting in 2014, but the net impact of the ACA to providers revenue should still be positive and allow room for reductions in unit costs. Therefore, total statewide provider revenue is estimated to increase significantly if there aren t changes to provider unit fees. To the extent this excess cost is driven by people seeking services who did not seek it before, changes will likely not drive down provider fees. To the extent this excess cost is driven by reimbursement for services that would have been provided without insurance coverage (uncompensated care), the increased revenue may allow providers to decrease unit costs. However, negotiations between providers and health insurance companies are complex, market specific and therefore difficult to predict. Another relevant initiative related to provider revenue and delivery system structure is the Medicaid Coordinated Care Organization (CCO) program that was recently developed by the Oregon Health Authority (OHA). The CCO program requires current health plans and provider systems to move towards a more integrated approach to care delivery. As part of this program, OHA is expecting significant savings. These new CCO organizations may help transform the market and create opportunities to leverage the new delivery models to the commercial marketplace. 2.7 How to Reduce Churning between the Individual Exchange and Medicaid? With the ACA-defined income corridors for Medicaid and Exchange APTC/CSR eligibility, there are estimates that up to 35% of enrollees in Medicaid (0-133% FPL) and recipients of high subsidy Premium Tax Credits ( PTCs ) and Cost Sharing Reduction Subsidies ( CSRS ) ( % FPL) will switch eligibility category every six months, and up to 50% of enrollees every 12 months. 11 Similar findings were reached in analysis of income and participation simulations conducted by Alex Graves, PhD at Vanderbilt, with the assistance of Jonathan Gruber at MIT. 12 The Commonwealth Care program (adults only, <300% FPL) in the Massachusetts Health Connector experiences 5% churn every month, a substantial portion of which is with MassHealth (Medicaid) See Sommers & Rosenbaum, Issues In Health Reform: How Changes In Eligibility May Move Millions Back And Forth Between Medicaid And Insurance Exchanges, Health Affairs, Vol. 30, No. 2, February, 2011; See also 12 Graves, Tabulations of the Survey of Income and Program Participation, with computing support and consultation from Jonathan Gruber, Ph.D., Professor of Economics at MIT. 13 Massachusetts Health Connector, Board of Directors, quarterly report July 31, 2012 Page 41

45 A change in eligibility can often mean a switch in coverage and/or health plans. The problems associated with discontinuity of coverage include: Risk of gaps in coverage, resulting in financial hardship for non-covered services and/or delayed service delivery; Disruption to patient-provider relationships; Discontinuity of care for chronic conditions and/or undermining patient-centered medical homes, with associated quality and cost issues; Increased Member cost-sharing, as change in plan enrollment can restart the clock on member cost sharing (e.g. deductibles). Member distress, inconvenience, and confusion, which can exacerbate anxiety-affected medical conditions and compromise access; Increased administrative expenses for the state and health plans; Reduced investments by the State and health plans in population health management as high churn diminishes plan opportunity for return on investment. Policy Options We describe below a menu of options for mitigating discontinuity of coverage and disruption of care inherent in churn. These options fall into three categories. One option is to discourage churn by guaranteeing eligibility over time. A second set of options is designed to retain enrollees in the same health plans, as they move from one coverage program to another. Because most of Oregon s Medicaid and CHIP enrollment on the Oregon Health Plan (OHP) is served by health plans (MMCOs), to the extent that the MMCOs also participate in ORHIX, there is at least the opportunity to maintain continuous enrollment in the same plan as eligibility changes occur. A third set of options address steps that OHP, ORHIX, QHPs, MMCOs and Navigators can take to ease transition, even if enrollees change plans as they change eligibility. Guarantee Continuous Enrollment: One way that Oregon might address the discontinuities inherent in churn is to minimize it by extending eligibility for the Oregon Health Plan (OHP). That is, Oregon could provide 12-month continuous eligibility and guaranteed enrollment for (non-abd) OHP enrollees. Adults (19-64) who are determined eligible for Medicaid (based on current income) would be guaranteed enrollment for a full 12 months, regardless of any subsequent changes during that period in their income or family composition. Federal rules prevent a parallel provision guaranteeing 1-year eligibility for subsidies in the exchange, so the cost of guaranteeing eligibility for 12 months would be exacerbated by the asymmetry of this rule in increasing OHP s total enrollment. July 31, 2012 Page 42

46 Temporary Continuation of Medicaid Coverage: Short of 12-month eligibility, OHC might arrange for termination of Medicaid/CHIP eligibility at month s end, so that former OHC enrollees can pick up coverage in the exchange at the start of the next month without a lapse in overage. Encourage MMCO Participation in ORHIX: Assuming other quality and licensure standards are met, OHP could encourage, or even require, that MMCOs become commercially licensed and participate in ORHIX, at least for individual coverage, if not SHOP. Oregon s Insurance Department might also grant provisional licensure, as a step toward full licensure of such MMCOs, and/or for those MMCOs that cannot meet reserve requirements, to allow them to participate on a limited basis in ORHIX (see Temporary Continuation below). In its certification criteria and process, ORHIX should encourage MMCOs to participate as QHPs. This will facilitate seamless transition by giving members multiple QHP options to address transition risk. Encourage MMCO & QHP Outreach to At Risk Members: ORHIX and OHP could encourage MMCOs to reach out to members at risk of transition. Some states significantly restrict MMCO marketing efforts and the information that can be provided to Medicaid beneficiaries, either in connection with MMCO selection or with ongoing eligibility determinations. Just as an applicant for tax credits (APTCs) and QHP enrollment may benefit from information about QHP selection, MMCOs could be permitted to provide information to OHP beneficiaries about transition risk and options upon transition. Temporary Continuation of MMCO Availability in ORHIX: ORHIX might arrange for MMCOs not otherwise participating in the Exchange to participate on a limited basis i.e., for temporary coverage of existing enrollees who lose Medicaid eligibility and gain eligibility for APTCs in the Exchange, at or below 200% of FPL. Such people often have depleted resources and may well return to Medicaid. Because of the relatively low cost of MMCOs, this alternative could afford continued free or nearly-free coverage up to about 200% FPL (i.e., if the tax credit is calculated based on a commercial QHP). This option would require some sort of proto-commercial licensure for MMCOs, and should be restricted to MMCOs which meet all of Oregon s commercial licensure standards, except for reserve requirements. (All other MMCOs should be encouraged to participate fully in the individual exchange.) Enrollment would be only through ORHIX, and should end (a) for enrollees whose income rises above 200% FPL, or (b) at the exchange s annual open enrollment, so as to provide temporary accommodation, but not encourage MMCOs to remain outside full participation in the individual exchange. Auto-Enrollment: A Medicaid beneficiary who is determined ineligible for Medicaid during the course of a calendar year is eligible for Special Enrollment in the Exchange. (45 CFR (b)(2)(ii)). Coverage is to be effective on the first day of the following month. This assumes, however, that a qualified individual will file an application and select a QHP within 60 days of the date of the triggering event (i.e. loss of July 31, 2012 Page 43

47 Medicaid eligibility). 45 CFR (c). Any delay in the determination of eligibility could result in a gap in coverage. 14 Exchanges may automatically enroll qualified individuals for good cause. 45 CFR (g). One such cause could be the circumstance of ineligibility for Medicaid, permitting auto-enrollment in the same (or controlled) plan on the Exchange. ORHIX adoption of auto-enrollment in this circumstance would permit the Exchange to auto-enroll a member in the same or related plan to prevent a gap in coverage and maintain continuity of provider relationships and care. Continuity of Coverage and Care: A number of requirements could be placed on QHPs to reduce transition discontinuities. As many of those churning above or below 138% FPL will be eligible for Cost- Sharing Reduction subsidies (<250% FPL) in the exchange, QHPs should be encouraged to eliminate or reduce their deductibles before reducing any other cost-sharing in order to meet the AV standards of 94%, 87% and 73%; this will minimize the disruption of coverage due to a new exchange enrollee encountering a deductible mid-year. MMCOs participating as QHPs should be encouraged to maintain consistent care management practices and personnel for enrollees as they transition, and should offer the same providers in OHP and ORHIX. And OHP and ORHIX should offer the same provider-finder tools that denote providers that participate in Medicaid and QHPs, to allow beneficiaries to look for plans that will allow them to retain their clinical providers as they change eligibility. Exchange and Medicaid Website Notations: Both OHP and ORHIX should educate enrollees at high transition risk (income < 200% FPL or recent Medicaid participation) about ways to reduce or eliminate discontinuities. For example, OHP and ORHIX could provide consumer decision support aides that describe the risk of transition upon change of eligibility and denote plans that participate in both programs. ORHIX and OHP webpages describing these plans might also include links to the plans provider directories, and such plans might be permitted to denote providers participating in Medicaid, CHIP and the exchange. With this information, a consumer could determine not only that the plan participates in OHP and ORHIX, but the member can confirm that their providers also participate in both programs. Navigator/Broker Selection, Training and Counseling Requirements: Navigator and broker training will be critical to providing consumers with sound guidance about selection of a QHP, including transition risk. Navigators could be selected on criteria that relate to serving Medicaid/CHIP and the exchange, 14 Medicaid uses a point in time income eligibility process, 42 CFR (h)(1), whereas Exchanges use an annual income eligibility process based on the prior year tax return, reconciled at year end based on actual MAGI. State Medicaid Agencies that are not receiving 100% FMAP for a Medicaid beneficiary have an incentive to qualify an individual for Exchange PTCs/CSRSs upon change in current income. A change in current income is almost assured to be reflected in inconsistent data between most recent tax filing and current income or change in circumstance like a job, increasing the risk of delay in determining eligibility. July 31, 2012 Page 44

48 including the requirement to serve all subsidized coverage programs. Navigators/brokers should obtain training and provide counseling to members earning below 201% of FPL, when selecting a QHP, about transition risk upon change in eligibility. This requirement would encourage individuals at high risk of transition to be knowledgeable about the risks at the time of QHP selection. Best practices counseling guides and performance assessment for Navigators might include counseling on transition risk for vulnerable enrollees. July 31, 2012 Page 45

49 3. ANALYSIS OF THE OREGON SMALL GROUP MARKET 3.1 Summary of Small Group Analysis In summary, we expect the following changes in the small group market in 2014 as compared to 2011 as a result of ACA provisions: Compression of rates caused by rating restrictions due to the introduction of new plan design requirements and also due to the elimination of underwriting factors based on employee health status, employee participation levels, and employer contribution levels. The highest amount of rate compression will happen for the smallest groups with the largest of the small groups incurring fairly insignificant change. Overall increase of 3% to small group premiums due to the ACA with a range of possible outcomes ranging from -5% to +14%. This range does not include the impact of the ACA requirements that have already gone into effect. The CBO estimates that the overall effect of the law on premiums for companies with fewer than 50 workers would range from an increase of 1% to a decrease of 2% in 2016, relative to current law. This is without the effects of the small business tax credit, which the CBO estimates would further reduce premiums by 8% to 11% for eligible firms. Insignificant impacts to the small group market size and product mix except for the addition of small groups currently in associations. We have not included a full analysis of the impact of expanded self-insurance options on the small group market post ACA. Anecdotal information suggests that the number of small employers that are switching from fully insured to self-insured is increasing, and reinsurers are targeting this market for further expansion. It is our understanding, however, that per ORS , Oregon prohibits stop loss policies from being sold to small employers. This effectively requires small groups to choose a fullyinsured option if they offer employer sponsored insurance. 3.2 Current Rating and Underwriting Rules in Oregon The current rating and underwriting rules provide an important context for analyzing the premium impacts and selection issues under the ACA since underwriting and rating practices are, at least partially, designed to mitigate adverse selection. Current regulations in these areas are intended to provide particular consumer protections while allowing health insurers to protect themselves against adverse selection. Regulations vary considerably by state. Oregon has somewhat atypical small group rating rules allowing for rate variation for variables such as participation and contribution. In some ways, however, it is similar to other states allowing rates to vary by age, duration, geographic area and member health status. Oregon currently defines small employer groups as those with 2 to 50 employees for rating and risk pooling purposes. The State requires the guarantee issue of insurance to small employers, meaning that July 31, 2012 Page 46

50 small employers cannot be turned down or cancelled because of the health conditions of their employees. However, insurers can require small employers to meet participation and contribution requirements. This means that insurers may require that a certain percentage of the employers within the group take coverage. They may also require that the employer pay a certain portion of the premium. If they do not, the small group coverage can be denied or terminated. Insurers are required to set premiums based on adjusted community rating, and can vary premiums based on age, geographic area, family composition, duration with insurer, employer contribution level, employee participation level, expected claims, tobacco use by employees and wellness programs. Specific rules regarding the adjusted community rating include: The factors based on employer contributions or employee participation may vary with the size of the employer. All other factors must be applied in the same actuarially sound way to all small employer health benefit plans. The adjustment for expected claims experience may not exceed five percent of the annual premium rate otherwise payable by the small employer. The premium rates charged during a rating period for health benefit plans issued to small employers may not vary from the geographic average rate by more than 50 percent. 3.3 Current Coverage and Costs in the Oregon Small Group Market Data Received The analysis was based on data provided by the insurers and the State. This information includes but is not limited to: Detailed benefit plan information for plans representing at least 80% of the insurer s small group book of business. The detailed information includes: o 2011 earned premiums, allowed and paid claims, and member months by benefit plan. The same data elements were also provided in aggregate for the balance of the remaining plans in the insurer s small group book of business. o High level cost sharing and covered services information for each benefit plan. o A summary of the variables utilized in rating along with detailed distribution data including 2011 member months and 2011 group months by rating factor. The following eight insurers provided information for the analysis: Health Net Kaiser Permanente ODS Health Plan, Inc. PacificSource July 31, 2012 Page 47

51 Premera/Lifewise Health Plan of Oregon Providence Health Plans Regence BCBSO United Healthcare All information provided by the carriers was for fully insured business. Overview of Current Small Group Market The following represents the insurance market in 2011 including sub markets that may be eligible to enter the small group exchange under ACA. Table 22: Carriers in the Oregon Small Group Market 2011 Average Members Carrier Small Assoc Assoc * Group < Total % of Total Health Net 36,307 10, ,168 49,060 13% Kaiser 32, , ,699 14% ODS 8, , ,131 4% PacificSource 34,616 11,095 11, ,931 16% Premera/Lifewise 7, , ,150 4% Providence 29,909 3,974 21, ,708 15% Regence 48,842 7,112 19,865 8,553 84,371 23% United 10, , ,582 3% Other 15,744 2,495 6, ,639 7% Total 224,908 35,649 92,684 13, , % % of Total 61% 10% 25% 4% 100% Note: Does not include other potential Small Group market migration. * data not available for Health Net. As compared to national statistics, the Oregon small group market has higher average premiums; and Oregon small employers are about as likely to offer Employer Sponsored Insurance (ESI). The higher premiums may be a result of any number of differences between the national market and the Oregon market, including richer benefits, higher provider reimbursement rates, an older and/or sicker workforce, or higher administrative costs. July 31, 2012 Page 48

52 Table 23: Oregon Small Group Market Compared to US Description United States Oregon Average Monthly Premium 2010 [1][2] Single Coverage $426 $447 Family Coverage $1,117 $1,186 Percent of Private Sector Establishments that offer Health Insurance to Employees 2010 [3] Firms with Fewer than 50 Employees 39.2% 39.7% Firms with 50 Employees or More 96.4% 94.7% [1] Source: AHIP Center for Policy and Research. Small Group Health Insurance in 2010: A Comprehensive Survey of Premiums, Product Choices, and Benefits, and statehealthfacts.org. Table 5 [2] Source: [3] Source: Agency for Healthcare Research and Quality, Center for Cost and Financing Studies Medical Expenditure Panel Survey - Insurance Component. Table II.A.2 When assessing the impact of ACA on the small group market, it is important to look at how the new law impacts various group sizes. Smaller group sizes behave more like individuals in that they have a wider variation of premiums, larger fluctuation of claims, and selection is a bigger factor. Larger group sizes, mainly because of the law of large numbers, tend to revert to the average of the market, with average age/gender factors and claims costs and therefore have less variation in costs. This variation in premium can be seen in the table below. The smaller group sizes have a wider variation in premium than the larger group sizes. Table 24: National Premium Variation by Group Size Small Group Market, 2010 Group Size 10th Percentile Mean 90th Percentile Ratio of 90th Percentile to 10th Percentile 10 or Fewer Employees $184 $446 $ : employees $185 $419 $ : employees $188 $406 $ :1 All Small Groups $181 $426 $ :1 Source: AHIP Center for Policy and Research. Small Group Health Insurance in 2010: A Comprehensive Survey of Premiums, Product Choices, and Benefits Along with this idea of keeping track of impacts by group size, it is also important to understand the difference between the distribution of group counts versus the distribution of members by group size. July 31, 2012 Page 49

53 Generally, the distribution of groups is not the same as the distribution of membership. Oregon is not an exception in this regard. The table below shows the Oregon distribution of groups and members. Table 25: 2011 Distribution of Groups and Distribution of Members Oregon Distribution of Groups Distribution of Members 10 or Fewer Employees 85% 52% Employees 11% 29% Employees 4% 19% The majority (85%) of the small groups have ten or fewer employees. However, these groups comprise only 52% of the small group market s membership. The average employer in the Oregon small group market covers 5.4 employees and has 9.1 members. The percent of employers offering coverage also varies by group size. In general, the smaller the group size, the less likely employers are to offer coverage. The table below shows the percent of employers offering coverage by group size for both Oregon and the nation in total. Table 26: Percent of Employers Offering Coverage, By Group Size Group Size National Oregon Less than 10 Employees 31.8% 34.2% employees 60.9% 56.5% employees 80.6% 81.7% Less than 50 employees 39.2% 39.7% 50 or more employees 96.4% 94.7% Source: Agency for Healthcare Research and Quality, Center for Cost and Financing Studies Medical Expenditure Panel Survey - Insurance Component. Table II.A.2. Of the employers that offer coverage, a smaller percentage of small employer groups chose to selfinsure. The table below shows the percent of covered enrollees that are enrolled in self-insured plans. The percent of group enrollees covered by self-insured plans in Oregon appears to be fairly consistent with the national level, but it is probable that most of these groups are association groups since Oregon prohibits stop loss policies from being sold to small employers. July 31, 2012 Page 50

54 Table 27: Percent of Covered Enrollees in Self-Funded Plans Group Size National Oregon Less than 50 employees 12.5% 14.8% 50 or more employees 67.5% 67.0% Source: Agency for Healthcare Research and Quality, Center for Cost and Financing Studies Medical Expenditure Panel Survey - Insurance Component. Table II.B.2.b.(1) Oregon Number does not meet standard of reliability or precision. Loss Ratios and Actuarial Values of Benefit Plans in the Small Group Market The loss ratios of seven of the eight small group insurers are consistent and range between 76% and 84%. One insurer has a loss ratio of 67%. The overall average loss ratio across all eight insurers is 80%. The AV reflects the relative richness of the benefit and is calculated by dividing the claims cost by the allowed amounts under the benefit plan. An AV of 100% means the benefit plan provides 100% coverage with no cost sharing. The average AVs by carrier in Oregon are surprisingly varied, with one carrier having an average AV of 66% and another carrier having an average AV of 87%. Different AV levels can indicate differences between the carriers in the risk of the underlying populations. This difference in AVs by carrier is somewhat correlated with the average premium by carrier, as would be expected. Table 28: 2011 Average AV and Average PMPM Revenue by Carrier Carrier Average Actuarial Value Average PMPM Revenue 1 66% $ % $ % $ % $ % $ % $ % $ % $340 July 31, 2012 Page 51

55 Figure 5: 2011 Average AV and Average PMPM Revenue by Carrier Average PMPM Revenue vs. Average Actuarial Value (excludes high and low AV) Average PMPM Revenue % 65% 70% 75% 80% 85% 90% Average Actuarial Value Current Rating Practices Per statute, carriers may vary rates by age, geographic area, family composition, duration with insurer, employer contribution level, employee participation level, expected claims, tobacco use by employees, and wellness programs. We were able to review rating factor distributions for most carriers included in the analysis. The table below summarizes which carriers utilize each of the allowable rating variables. Table 29: Allowable Rating Variables by Carriers Duration with Insurer Expected Claims (+/-5%) Carrier Age Area Contract Tier Employer Contribution Employee Participation Tobacco Use Carrier 1 Yes Yes Yes No No Yes Yes Yes Yes Carrier 2 Yes No Yes No No Yes Yes No No Carrier 3 Yes Yes Yes No Yes Yes No No No Carrier 4 Yes Yes Yes No Yes Yes Yes No No Carrier 5 Yes Yes Yes No Yes Yes Yes No No Carrier 6 Yes Yes Yes No No No Yes No No Carrier 7 Yes Yes Yes Yes Yes Yes Yes No No Carrier 8 Yes Yes Yes No No Yes Yes No No ACA allowable? Limited to 3:1 Yes Yes No No No No Limited to load of 50% Wellness Programs No July 31, 2012 Page 52

56 Age. Age is an allowed case characteristic. All of the carriers have age ratios greater than 3:1, the allowable limit under the ACA. They range from around 3.8:1 to 5.9:1. Area. Area is an allowed case characteristic. Seven of the eight carriers use area as a rating variable. They assign a unique factor to each of seven distinct geographic areas within the state of Oregon. Most carriers have a spread of approximately 15% between the lowest area factor and the highest area factor. One carrier, however, has a spread of over 50% from the lowest to the highest factor. Duration with Insurer. While duration with insurer is an allowable rating factor, it is only used by one of the eight insurers. Employer Contribution. Oregon allows insurers to rate based on the level of employer contribution and half of the eight carriers utilize this rating variable. In general, carriers rate up when employer contribution levels are low. While the state allows such factors to vary by group size, only one insurer does this in practice. Employer Participation. All but one carrier utilizes the allowable rating variable of employee participation level. The majority of these carriers vary the rating factors by group size, as permitted by the state of Oregon. These factors generally decrease as participation levels increase. They also decrease as group size increases. Interestingly, varying employer participation factors (and employer contribution factors) by group size also results in a group size rating factor. Expected Claims. Health status, or expected claims level, is an allowed case characteristic and seven of the eight carriers make use of it in rating. The variation of health status rating is currently limited to +/- 5%. Tobacco Use and Wellness. Tobacco use and wellness programs are permissible rating characteristics in Oregon, but only one of the eight carriers apply them in practice. Benefit plan. Various plan designs are currently available in the market. Table 30 below shows the distribution of small group membership by deductible and in-network plan coinsurance. July 31, 2012 Page 53

57 Table 30: Percent of Membership Enrolled in Plans In-network Deductible In-network Plan Coinsurance % of Membership $ % 23.7% $501-1,000 80% 16.9% $1,001-2,000 80% 13.7% $1,001-2,000 70% 7.9% $2,001-3,000 80% 6.6% $2,001-3,000 70% 6.1% $ % 4.6% $4,001-5,000 70% 4.4% $501-1,000 70% 3.9% $4,001-5,000 80% 3.8% $ % 1.5% $ % 1.1% $5,001-7,500 70% 1.0% Other Other 4.8% 3.4 ACA Impact on Small Group Market Summary In studying the impact that the ACA will have on the small group market, there are many factors to consider. In this section, we review provisions of the ACA both inside and outside the Small Business Options Program (SHOP) exchange and their impact on: Premium rates Product offering Market size The small group market post ACA has many segments: the groups accessing insurance coverage through the SHOP exchange, those fully insured outside of the exchange and grandfathered, those fully insured outside of the exchange and not grandfathered, and those that choose to self-insure. The ACA impacts all fully insured non-grandfathered plans, has limited impact to grandfathered plans, and has little to no impact to the self-insured market. July 31, 2012 Page 54

58 Premium Impact This section addresses the impact that the ACA will have on fully insured non-grandfathered plans. Underwriting Rules The general guarantee issue requirements in the ACA (for children in 2010 and for all ages in 2014) are not expected to have significant effect on the small group market rates given the current guarantee issue requirements in Oregon s small employer insurance market. The expansion of coverage to dependents to age 26 (implemented in the fall of 2010) was expected to have an overall increase to small group employer rates due to the additional members covered, but an overall decrease to the PMPM rates (as the average risk of the additional members is less than the small group market as a whole). The experience we received does not include the data necessary to evaluate the impact of this change. We expect the average PMPM premiums to have decreased between 0.5% and 1%. New Plan Design Restrictions Starting in 2010, the ACA incorporated the following benefit design restrictions: 1. Removal of lifetime limits and restricted annual maximums % coverage of preventive care services 3. Coverage of emergency services 4. Choice of provider Again, the data do not explicitly indicate the rate change due to these changes. We expect the average PMPM premiums to have increased by 1-2% as a result of these changes. Starting in 2014, the ACA has the following main provisions regarding plan design: 1. Prohibition on annual limits; 2. Essential health benefits requirements; 3. Minimum actuarial value requirements; and 4. Limits on member cost-sharing, including maximum annual deductibles of $2,000 (individual) and $4,000 (family). Starting in 2014, all health plans sold through the exchange and in the individual and small group markets will be required to cover the essential health benefits. The ACA defines essential health benefits to include at least the following general categories and the items and services covered within the categories: ambulatory patient services; emergency services; hospitalization; maternity and newborn care; mental health and substance use disorder services, including behavioral health treatment; July 31, 2012 Page 55

59 prescription drugs; rehabilitative and habilitative services and devices; laboratory services; preventive and wellness services and chronic disease management; and pediatric services, including oral and vision care. 15 In addition, health plans are required to offer coverage in one of four levels (platinum, gold, silver and bronze), based on the relative actuarial value (AV), which is a ratio of the expected paid amount to the expected allowed amount of costs in a plan. An AV of 100% would provide full coverage with no costsharing. Table 31: Actuarial Value by ACA Plan Level Plan Design Names Actuarial Value Bronze 60% Silver 70% Gold 80% Platinum 90% Regulations clarifying how these minimum values will be set by plan level have not yet been released. However, how or if these benefits are mandated outside of the exchange is not yet clear. The ACA requires that each health insurer that offers health insurance coverage in the individual or small group market shall ensure that such coverage includes the essential health benefits package. The scope of this requirement continues to be debated, and may imply that all fully insured non-grandfathered health plans both inside and outside of SHOP all need to comply with the benefit minimums as well as be equivalent to one of four specific plans as defined by the actuarial value. The Congressional Research Service has indicated that both the essential health benefits and standardized benefit packages will be required both inside and outside the exchange. 16 The ACA limits cost-sharing in health plans by setting maximum deductibles and maximum out of pocket amounts. Deductibles for group coverage can be no more than $2,000 for a single policy or $4,000 for a family policy; and total out-of-pocket are limited to the amounts set each year by the Internal Revenue Service for high deductible health plans, which in CY 2011 are $5,950 for individual coverage and $11,900 for a family policy. The deductible limits do not apply to coverage purchased in the individual market. Our analysis of the current small group plan designs offered in Oregon indicates: 15 Public Laws & PPACA Requirements for Offering Health Insurance Inside Versus Outside an Exchange, Congressional Research Service, June 1, July 31, 2012 Page 56

60 A majority of the small group membership is covered by plan designs that are likely to be compliant with the plan design requirements of the ACA. Approximately 6% of the membership is in plans that will require an increase in coverage of approximately 10% due to having benefit plans with actuarial values that are less than the minimum 60% required for the bronze plan. This will result in an overall 0.2% increase in the average premiums for the small group market. Approximately 25% of the current membership is in benefit plans that have deductibles exceeding the ACA allowable limit and approximately 18% of the current membership is in benefit plans that have out of pocket maximums exceeding the ACA allowable limit. Nearly one quarter of these non-compliant benefit plans also have an actuarial value below 60% (addressed above) and will have to increase coverage anyway to meet the minimum threshold. Because the actuarial values for most plans are well above the 60% minimum, small employer insurers will likely transition employer groups that are not compliant with ACA benefit requirements to new plans that meet the benefit requirements but have similar costs, keeping rates similar to current rates. Therefore, we project no rate impact for groups that have plan designs that currently have AVs above the Bronze level, even if some of the benefits underlying these designs are not compliant. New Rating Rules Starting in 2014, the ACA introduces many new rating restrictions for non-grandfathered, fully insured health plans. The main rating rules to discuss include: Adjusted Community Rating Age and gender rating restrictions Elimination of underwriting factors Family composition requirements Rating area- state determined, reviewed and approved by the secretary Tobacco use (1.5:1) Many of the carriers currently offer products to small groups through associations and these groups will also be affected by the rating rules. Under current law, association groups are permitted to use experience rating techniques which will no longer be permissible under the ACA. As a result, association groups may experience more significant changes to premiums post ACA than their non-association counterparts. Like small group, most association < 51 benefit plans are already compliant with the ACA standards. About 12% of current association < 51 members are in plans with deductibles exceeding the post ACA allowable limit and approximately 7% of members are in plans with maximum out of pocket values exceeding the ACA allowable limit. July 31, 2012 Page 57

61 Most carriers do not track rating variables for their association data so we were unable to calculate the specific premium impacts that will result from each ACA rule change. All premium impacts are only in reference to the non-association market unless indicated otherwise. Overall, it is not expected that any one of these rating rules will have a significant impact on the average small group premiums as long as the same pool of groups remain in the small group market. However, depending on the characteristics of the group, some employers will see premium increases and some will see premium decreases. The adjusted community rating rule requires that issuers must consider all enrollees of their small group plans part of a single risk pool, including enrollees in small group plans both inside and outside the exchange. The premiums for each group in the pool will be based on the average experience of the entire pool, adjusted for only the rating variation factors that the ACA permits. The age and gender rating restrictions introduced by the ACA require that the highest adult rate be no more than three times the cost of the lowest adult rate and that gender cannot be used as a rating factor. Similar to the individual market analysis, groups with the youngest employees will see a rate increase and groups with the oldest employees will see rate decreases. However, the small group market rates will behave somewhat differently than the individual market in that group rates generally reflect the average rate of all employees. Therefore in order for an average group rate to change, the majority of the employees of the group would have to be at either the very youngest or the very oldest ages. Due to the law of large numbers, the smallest groups would have the highest likelihood of being affected by a change in average premium due to a change in the age gender slope of the rates. One of the major concerning impacts of the ACA for small groups is the elimination of underwriting factors. Currently, Oregon requires the variation of underwriting factors to be no greater than +/-5%. This variation will be preempted by the ACA in Underwriting factors allow insurers to vary premiums consistently with the expected costs of the groups. All groups with underwriting factors different from the average will be required to be adjusted back to the average premium rate. Based on information provided by carriers in the Oregon small group market, the major carriers all have average underwriting factors by carrier of around 1.0, with the average underwriting factor ranging from 0.97 to The table below shows the impact of eliminating the underwriting factors. For example, 3.3% of the groups will receive a 3% to 4% increase in rates. July 31, 2012 Page 58

62 Table 32: Rate Change Distribution Due to Removal of Underwriting Factors Rate Change Groups Members -5% to -4% 8.9% 6.9% -4% to -3% 16.7% 16.3% -3% to -2% 6.1% 8.4% -2% to -1% 2.3% 2.5% -1% to 0% 1.7% 3.7% 0% to 1% 23.9% 12.3% 1% to 2% 24.4% 22.6% 2% to 3% 12.7% 22.1% 3% to 4% 3.3% 5.3% The above distribution is further split into group size in the table below. Table 33: Rate Change Distribution by Group Size -5% to -4% -4% to -3% Percentages Based on Group Counts -3% to -2% to -1% 0% to 1% to -2% -1% to 0% 1% 2% 2% to 3% 3% to 4% Total 2 to 5 Employees 7.3% 13.3% 2.7% 1.5% 0.5% 22.0% 18.0% 6.2% 1.2% 72.9% 6 to 10 Employees 1.1% 1.7% 1.6% 0.4% 0.6% 1.6% 3.8% 3.2% 1.0% 14.9% 11 to 20 Employees 0.4% 1.0% 1.3% 0.2% 0.3% 0.3% 1.9% 1.9% 0.7% 8.1% 21 to 30 Employees 0.1% 0.4% 0.4% 0.1% 0.2% 0.1% 0.4% 0.8% 0.3% 2.7% 31 to 40 Employees 0.0% 0.1% 0.1% 0.0% 0.1% 0.0% 0.2% 0.4% 0.1% 1.0% 41 to 50 Employees 0.0% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.2% 0.0% 0.4% Total 8.9% 16.7% 6.1% 2.3% 1.7% 23.9% 24.4% 12.7% 3.3% 100.0% As discussed above, the table also shows that 3.3% of the groups will receive a 3% to 4% increase in rates, and nearly one third of these groups or 1.2% out of 3.3% - are groups of two to five employees. In general, the smaller group sizes will see greater changes in premium (both up and down) due to the elimination of underwriting factors. Rating by family composition (employee only, employee plus spouse, employee plus children, employee plus family) under the ACA appears to be consistent with current rating practices. However, the July 31, 2012 Page 59

63 proposed rules are requesting comment as to what restrictions should be established around the tier structuring. Based on the guidance to date, we do not believe that the tier structuring rules will impact rates. The ACA allows for rating by geography, as do state regulations. Based on the data reviewed, seven of the eight carriers use this variable when setting rates. This rating restriction will have no impact on small group premiums. Another rating factor that the ACA allows is tobacco use. Rates for consumers can be increased up to 50% if the consumer uses tobacco. Currently, Oregon allows insurers to rate based on tobacco use, but only one of the eight carriers for which we received data actually utilize this rating variable. This rating restriction will have no impact on small group premiums since the only carrier utilizing this rating characteristic already falls within the ACA permissible range. However, if issuers begin to use a tobacco rating factor in a manner different that their current rating process, it could affect small group premiums. As for the other rating variables, carriers will not be permitted to vary rates based on employer contribution or employee participation under the ACA. Half of the eight carriers utilize employer contribution as a rating characteristic currently. While the elimination of this rating variable will have no impact on rates overall, individual groups could experience premium impacts anywhere from -7% to +4%. The majority of the carriers for which we received data apply employee participation factors in rating currently. The elimination of this rating variable will not impact small group premiums at the market level, but individual groups could experience premium impacts anywhere from -14% to +8%. The ACA no longer allows rating by class, industry factor, or other factors such as the number of selected health plans by group. Based on the data we reviewed, we do not believe that these rating factors are used in Oregon, and therefore believe the elimination of these rating factors will have minimal impact on the small group rates. New MLR requirements The new medical loss ratio requirement, effective in 2011, requires that small group carriers maintain a loss ratio of at least 80%. Any carriers with loss ratios below 80% are required to rebate the difference in premiums back to the members. As part of this analysis, we were provided incurred claims and earned premiums for 2011 small group lines of business. After adjusting for taxes and assessments and applying ACA allowable credibility and special circumstances adjustments, only one of the eight carriers has a small group loss ratio below the 80% minimum requirement. Therefore, we expect the minimum loss ratio requirement to decrease premiums by a mere 0.1%. July 31, 2012 Page 60

64 The Health Insurance Provider Fee As stated under the individual section, Wakely developed estimates of the insurer fee and its impact on premiums. Because the average impact of the insurer fee on premium rates depends on the proportion of non-profit and tax exempt premiums, and Oregon has a high proportion of non-profits, the impact of the insurer fee on premium rates in Oregon is less than the national average. The following table shows the best estimate impact of adjusting the national average values to Oregon: Table 34: Wakely Estimates of Oregon Premium Impact by Year Provider Fee Insurer Type Distribution of Members For Profit 34.8% 1.96% 2.53% 2.34% Not for Profit 65.2% 0.00% 0.00% 0.00% Total 100.0% 0.68% 0.88% 0.81% Note that if the distribution of members is adjusted for 2016 to include the members the total impact decreases to 0.73% since the members have a higher concentration of members in not for profit insurers. Tax Credits From 2010 through 2013, until the SHOP exchanges are set up, businesses with 10 or fewer full-timeequivalent employees earning less than $25,000 a year on average will be eligible for a tax credit of 35% of health insurance costs. (Companies with between 11 and 25 workers and an average wage of up to $50,000 are eligible for partial credits.) The tax credit will remain in place, increasing to 50% of costs, for the first two years a company buys insurance through its state s SHOP. The Congressional Budget Office predicts that the tax credit will affect about 12% of individuals covered via the small-group insurance market, lowering their cost of insurance by between 8% and 11%. We anticipate tax credits will offset approximately 1% of the total premium paid by small group employers. It s important to note, however, that small employers can only claim the credit for 2010 through 2013 and for two additional years beginning in 2014, which limits the long term effect of the employer tax credit July 31, 2012 Page 61

65 Other Items Impacting the Overall Morbidity of the Risk Pool While the specific rating provisions described above will have varying impacts to existing groups based on the demographic characteristics and existing benefit coverage of each group, considerable thought and time needs to be spent on understanding the entrants and exits of various employer groups to the market. According to 2010 Medical Expenditure Panel Survey (MEPS) data, approximately 40% of Oregon small employers offer ESI. Of those employees that are offered ESI by small employers, only 61% enroll in coverage. 18 Key questions remain around whether small employers will be more or less likely to purchase insurance in 2014, whether employees will be more or less likely to sign up for coverage, and the relative health risk of the currently uninsured consumers relative to the existing small group market. One of the main exchange issues states are wrestling with is the mitigation of selection inside the fully insured pool and how to structure the SHOP exchange so that it does not get adversely selected against and attracts a broad cross-section of healthy and sick individuals. CCIIO, in their guidance to states on exchanges, state Successful Exchanges will avoid adverse selection by ensuring that those who buy through the Exchange are a broad mix of the healthy and the less healthy. The tax credits, which can only be accessed through the Exchanges, and insurance reforms required by the Affordable Care Act will reduce the potential for adverse selection against the Exchange, but will not eliminate it. States have flexibility to provide consistent regulation inside and outside the Exchange, and to take additional action to prevent adverse selection under section 1311(e)(1)(B). The federal government will work with States to maximize State flexibility in this area. 19 Adverse selection in the context of the exchanges generally refers to individuals propensity, acting as direct purchasers, to make decisions that benefit themselves, to the detriment of the insurance market in general or to a specific insurance issuer. Adverse selection occurs when healthy people decide not to purchase insurance, or purchase the minimum coverage necessary or when sick individuals only purchase insurance when they know they will need it, or when sick individuals purchase policies that cover the maximum amount of their expected costs. Carriers can influence the selection process by offering only certain products or marketing to a select group of individuals. These actions can quickly multiply when they are happening simultaneously, creating a marketplace with escalating costs and decreasing participation. Avoiding and mitigating adverse selection is essential to the success of the exchange and the broader small group market, in order to keep coverage affordable for consumers, particularly those without access to subsidies July 31, 2012 Page 62

66 While the various ACA provisions protect the exchange, QHPs and issuers against adverse selection, there are a few provisions that increase the potential for adverse selection in the small group market. No health status based rating or underwriting allowed. Age rating allowed at a ratio of three to one and gender rating is not allowed. Elimination of employer contribution and employee participation as rating variables. Availability of plans outside the exchange, through grandfathered plans, self-insured options and potentially other innovations of the market On the other side of the selection issue, certain ACA and state provisions could attract a cross-section of risks into the small group market. Single small group risk pool Individual mandate Penalties for employer groups Some provisions of the ACA are unclear regarding whether they mitigate or create adverse selection issues. Employer tax credits. Subsidies for individuals. Risk adjustment. Several of these provisions are examined further in the following sections. Health Status and Age Rating Underwriting and unlimited age gender rating (items 1 and 2) in the current market both allow insurance issuers to closely match the premiums of the health plans to the expected medical costs. When these provisions are removed, the youngest and healthiest employer groups will receive increased premiums for the same benefits, while the oldest and sickest employer groups will receive decreased premiums for the same benefits. This creates incentive for the young and healthy groups to find other sources of insurance or drop coverage altogether if premiums become cost-prohibitive. Table 35 below illustrates three scenarios of how the removal of the lowest risk score groups will increase the overall rates of the small group pool. July 31, 2012 Page 63

67 Table 35: Scenarios on Removal of Lowest Risk Score Groups Rate Change -5% to -4% -4% to -3% -3% to -2% -2% to -1% -1% to 0% 0% to 1% 1% to 2% 2% to 3% 3% to 4% Percent of Members Average Underwriting Factor Scenario 1-1/2 of healthiest members leave Membership Leaving Market Remaining Distribution Average Underwriting Factor Scenario 2 - All healthiest members leave Membership Leaving Market Remaining Distribution Average Underwriting Factor Scenario 3 Worst Case Membership Leaving Market Remaining Distribution 7% % 8% % 10% % 18% % % 19% % 22% % 43% % % 10% % 12% % 22% % % 3% % 3% % 7% % % 4% % 5% % 10% % % 14% % 17% % 0% % % 26% % 31% % 0% % % 13% % 0% % 0% % % 3% % 0% % 0% Total 100% % % % Overall Premium Impact 0.4% 1.0% 3.0% The scenarios are as follows: Scenario 1: Half of the groups who would see a 2% to 4% increase due to the removal of the underwriting factor would leave the market. All other groups stay. Scenario 2: All of the groups who get a 2% to 4% increase due to the removal of the underwriting factor leave the market. All other groups stay. Scenario 3: All groups who get an increase due to the removal of the underwriting factor leave the market. All other groups stay. The range of these scenarios is a 0.4% to 3.0% increase in the average premium of the small group market due to the elimination of the expected claims rating variable. These results could be more significant to the extent that actual morbidity for the current small market exceeds +/-5% and to the extent that employer group behavior varies from the three scenarios laid out above. Our research of states that have moved to adjusted or modified community rating (where they have eliminated the flexibility of an underwriting factor) shows mixed results. An AHIP study of small group rates across multiple states shows that states that do not allow rates to vary by health status generally Average Underwriting Factor July 31, 2012 Page 64

68 have higher average rates. 20 However, detailed studies on the state experience in Pennsylvania, Connecticut, and New York indicate that groups have not significantly selected out of the market due to the removal of the underwriting factor from the allowed rating methodologies. 21 A similar study in Colorado indicates that the removal of the underwriting factor resulted in less availability of carriers in the market, and decline in small employer coverage, but not spiraling increases in the rates. 22 Estimating the impact of healthy small groups leaving the small group risk pool requires understanding where the groups would go for alternate coverage and potentially better rates. Options include staying in a grandfathered plan, entering the self-insured market, moving all the workers to the individual market by simply removing coverage, or potentially other non-regulated fully insured options like associations or Professional Employer Organizations (PEOs). Availability of Plans Inside and Outside the Exchange Health Plans outside the exchange may be grandfathered meaning that they are allowed to remain in place as long as they meet requirements for maintaining their pre-reform benefit levels, copays, contribution levels, and covered services. In addition, in order to maintain grandfathered status, employer groups must stay with the same plan design they were enrolled in as of March 23, Grandfathered plans are not subject to many ACA rules such as adjusted community rating, essential benefits and age gender restrictions. Grandfathered plans will not be available within the exchange, and may drive a different balance of risk between the population covered inside and outside the small group pool. The ACA does not require that grandfathered plans be included in the same pool as nongrandfathered fully insured plans sold in the small group market. The Blue Cross Blue Shield Association released projections that the majority of small employer group plans (size 3-99) will lose their grandfathered status by ; however, we do not have any specific Oregon information on this statistic. Self-funding options are attractive to healthy groups, but have generally not been accessed by the small group market historically. Only approximately 14.8% of employees in Oregon small groups with 50 or fewer employees are covered by a self-funded plan versus 67.0% above that level. 24 While many states (Oregon figure does not meet standard of reliability or precision) July 31, 2012 Page 65

69 are concerned about the expansion of self-insurance in the small employer market, our understanding is that per ORS (3), Oregon prohibits stop loss policies from being sold to small employers. This effectively requires small groups to choose a fully-insured option if they offer employer sponsored insurance. Promoting Cross Section of Risks The provisions below encourage a cross-section of risks in the small group market. Single small group risk pool Individual mandate Penalties for employer groups Although grandfathering and self-funding options will prevent some of the healthier groups from joining the small group risk pool, the ACA requires that all fully insured small group business sold either inside or outside of the exchange be included in the same risk pool and also the same risk adjustment calculation which will keep both rates inside and outside the SHOP on an even par. Furthermore, any QHP that is offered in the exchange must agree to charge the same premium outside the exchange. 25 However, the ACA does not prevent carriers from offering plans only outside the SHOP. The individual mandate and employer group penalties encourage groups with healthier employees to obtain small employer insurance. The participation of healthier employees in the market will help offset the increase in premiums, and will encourage participation by other healthy groups, preventing the situation where only groups with high cost members are purchasing insurance in the small group market and drive up costs. We note that the penalties for employer groups do not apply to employees with under 50 employees, and so will only impact the small group market once the small group definition is changed to include groups with up to 100 employees. The estimated amounts of increased membership in the small group market resulting from these provisions vary. See the Market Size Impact section which shows the estimates of newly offered ESI. Individual Premium Subsidies and Tax Credits of Offering ESI Coverage The ACA provides for premium tax credits for lower income individuals, which subsidize the cost of health coverage. As the subsidies are only available within the individual exchange and not through employer sponsored insurance, there is a consistent expectation across the research that the ESI market 25 Footnote: PPACA Requirements for Offering Health Insurance Inside Versus Outside an Exchange, Congressional Research Service July 31, 2012 Page 66

70 will lose the lower income workers (and their families) to the individual exchange. 26 It is less clear, however, whether these workers are higher or lower risk than the average ESI enrollment. To the extent that those needing subsidies are also higher than average risk, individual exchange adverse selection may occur and the ESI market will benefit. Because the low income individuals are generally younger, we believe it is more likely that those leaving the ESI market will actually be better than the average risk, leaving worse risk members in the small group ESI market. A research paper for Maine quantified the impact of the removal of the workers eligible for subsidies would increase the premiums in the small group ESI market by 6% to 7%. 27 Whether this will be the case in Oregon is not yet clear and warrants follow-up analysis. Given the uncertainty, for the purposes of this analysis, we assumed that any individual migration would have a similar risk as the current small group market. Tax advantages of offering coverage, penalties for employers with over 50 employees, the individual mandate, and removal of employer contribution requirements will mitigate the impact of the low income members leaving the ESI market and may encourage some employers who previously did not offer insurance to start offering ESI. Rand, through their micro simulation models, projects that the ACA will result in increased employer offer rates from 57% under the status quo to 80% for firms with 50 or fewer workers 90% to 98% for firms with workers 93% to 98% for firms with more than 100 workers. Already some insurers are seeing increased small employer enrollments, potentially due to the available tax subsidies. 28 This doesn t necessarily indicate that all employers in this group who were previously not providing coverage who join the small group market will be a good cross-section of risk, but from a small group perspective, increased numbers of employers joining the small group market likely increases the cross-section of risk in the pool. Risk Adjustment The risk adjustment mechanism is intended to help mitigate adverse selection by encouraging issuers to: 1. set premium rates based on average morbidity of the pool 2. offer benefit plans that attract higher risk, such as higher AV plans or more expanded networks 26 Avalere: The Affordable Care Act s Impact on Employer Sponsored Insurance: A Look at the Microsimulation Models and Other Analyses July 31, 2012 Page 67

71 Based on the risk adjustment mechanism, revenue for a carrier will either be increased or decreased based on whether the risk for a carrier is higher or lower respectively than the average after accounting for allowable rating factors (most notably age). The risk adjustment mechanism may help mitigate adverse selection among carriers, as additional revenue can be expected from the risk adjustment mechanism if higher than average risk membership enrolls. The additional revenues should curb the typical carrier s incentives to cherry pick the healthiest of employer groups when using adjusted community rating. Small Group and Association < 51 Merger The ACA will require association groups with fewer than 51 members to comply with the same rules and regulations applicable to the small group market. That is, the small group and association markets will merge. From a very simple perspective, merging the markets will equalize premiums. Therefore, if premiums are lower in the small group market prior to the merger, then small group premiums will increase and association < 51 premiums will decrease (or vice versa). The amount of the change in each market depends on the relative size of the markets prior to the merger. If the total market is dominated by small group, the change to association < 51 premiums could be substantial. The table below illustrates the post ACA allowed PMPMs for each of the submarkets along with the average 2011 membership for each. Current small group membership comprises the majority of total membership so the post ACA combined PMPM is weighted more heavily towards the current small group PMPM. Small group costs are lower than association < 51 costs so the merger would result in an increase of approximately 1% to small group premiums and a decrease of approximately 6% to association < 51 premiums. Table 36: Impact of Small Group and Association < 51 Merger Sub Market Post ACA Allowed PMPM Average 2011 Members Impact Small Group $ ,908 1% Association < 51 $467 35,649-6% Combined $ ,557 Product Offering Impact The main ACA provisions affecting the products offered, including the services and the cost-sharing provisions are: 1. Essential Health Benefits 2. Required Actuarial Values July 31, 2012 Page 68

72 In addition to covering the essential benefits, cost-sharing must be limited or conversely a minimum level of coverage must be provided on the essential benefits. This level of coverage is defined by the four levels of actuarial values and has various restrictions on out-of-pocket spending and deductibles. As insurers seek to differentiate themselves in the newly reformed market, one potential result is a change to the mix of PPO, HMO/POS, HSA/HDHP or indemnity offerings. In New York, when the small group rating moved to community rating, one of the changes in the market was that more HMO/POS products emerged. 29 An almost certain outcome of the ACA is that the small group market will result in greater homogeneity of the products offered from an AV standpoint. Employer Coverage Decisions In response to various ACA provisions and rating changes, insurers will make decisions to stay or leave the small group market. Generally, we do not see an exit of any major insurers from the market, and have some reason to believe that there may be some new entrants into the Oregon market. The following provisions of the ACA and the forthcoming regulations will impact how the market evolves. 1. Regional plans and regional definitions. It is not yet clear if QHPs will be able to operate in only specific regions (defined as a limited area within a state). 2. Rules in exchange /outside of exchange. Depending on how Oregon restricts benefit plans and rating outside of the exchange, whether small group and individual is pooled, and how soon groups up to size 100 are included in the exchange, some carriers might opt to only offer small group products outside of the exchange. 3. New types of health plans. With the possible introduction of co-ops, multi-state plans and health choice compacts, some opportunity exists for new insurers to enter the Oregon market. 4. Tax credit. The tax credit may cause small group employers to begin offering employer sponsored coverage, which has the potential to increase the overall size of the small group ESI market. Avalere has done a comparison of various micro-simulation models that project impacts on the employer sponsored market. Below is a table showing their summary July 31, 2012 Page 69

73 Source: Avalere: The Affordable Care Act s Impact on Employer Sponsored Insurance: A Look at the Micro simulation Models and Other Analyses While these models are not specific to the small group market, we believe there are some take-aways that are consistent across the models and applicable to the small group market: 1. The employer sponsored market will stay relatively stable, and 2. The main loss of coverage from employer groups are where the workers are eligible for the subsidies in the exchange and the employer does not offer affordable coverage. Forthcoming Regulations We are hopeful that many items that impact small group in the statute will become clearer with forthcoming regulations. Of particular interest are: Requirements of actuarial value determinations and de minimus rules Essential health benefits determinations. July 31, 2012 Page 70

74 Geographic rating requirements. How consumer choice in an employer sponsored coverage decision in the exchange will be allowed or restricted. Expanding Small Group Definition of 50 to 100 Effective for all plan years starting January 1, 2016 or after, the ACA requires the small group definition to be inclusive of all groups with up to 100 employees. However it allows the restriction of the small group definition to 50 employees for all plan years starting prior to that date. Note that this means that during 2016, there will be some groups of size employees that will be considered large group and some that will be considered small group, determined by the starting date of their plan year. Starting in 2017, all groups up to size 100 will be considered small group. The majority of the documentation published to date regarding this decision recommends that States continue to restrict the definition to 50 lives until statutorily required to expand. The rationale for this is risk mitigation. Businesses with 51 to 100 workers are more likely to have alternative coverage arrangements marketed to them, including self-insured plan arrangements including some sort of stop-loss reinsurance. Allowing businesses with employees into the small group market immediately could raise premiums because of adverse selection. Businesses with healthy workforces would choose to self-insure, while businesses with less healthy workforces would choose to take advantage of the non-health-rated coverage available through the small group market. As discussed above, we do not believe this risk applies to Oregon because the state prohibits stop loss policies from being sold to small employers. We assumed this regulation would extend to groups of size 51 to 100 if the definition was expanded. In addition to the adverse selection concern outlined above, some of the other considerations that will impact this decision include: All employer groups with more than 50 employees will face a penalty if they do not offer group insurance starting in The loss ratios, benefit designs, and administrative charges of the large group market compared to the small group market. Relative size of the market compared to the current small group market. All groups with more than 50 employees will have a financial incentive to offer coverage starting in This may cause employers that currently do not offer coverage to do so. Given that they will be just entering the insurance market, we believe such employer groups will be less sophisticated purchasers unlikely to enter into a self-insured or alternative arrangement and likely to enter the fully insured market. This fresh pool of members could be healthier than the average mix of membership for the small group market. July 31, 2012 Page 71

75 Generally, the loss ratios for the large group market are greater, causing lower premiums for larger groups compared to the small group market. This difference in administrative cost must be considered when combining the markets. Lastly, the relative size of the group market to the small group market is surprisingly small. Without econometric models, it is difficult to tell, but our best information based on combining state information about group sizes from the HRET KFF study as well as other data indicate that the group market is likely only between 25% and 50% of the existing small group market. At most, the groups of size will comprise only one-third of the small group market, and therefore would have somewhat limited impact to the overall small group risk pool. The results of this analysis hold true in Oregon about 29% of the total membership in groups of size 2 to 100 come from the 51 to 100 group size subset. Similar to the analysis done in merging the small group and individual markets, we have examined potential outcomes of expanding the small group market to 100 employees. One carrier did not provide data for the group market between 51 and 100 lives. We assumed that carrier s 51 to 100 group market behaves similarly to the general large group market data that we did receive. Based on the analysis summarized below, the average risk of the small group market would not be significantly affected by the addition of the group sizes Note that the Association 51 to 100 market was also included in this analysis. If the small group definition was increased to 100, these groups would also be subject to Small Group regulations. The table below shows that the average allowed PMPM in the expanded small group market (including association groups and groups of size 51 to 100) would be $435. This is the same as the current small group market (excluding association groups) and about 1% lower than the new small group market (including association groups of size < 51). This is partially due to the relative size of the < 51 market to the 51 to 100 market (as discussed above). Table 37: Impact of Increasing Small Group Definition to 100 Sub Market Post ACA Allowed PMPM Average 2011 Members Impact Small Group $ ,908 0% Association < 51 $467 35,649-7% New Small Group Market (with Association) $ ,557-1% 51 to 100 $428 92,684 2% Association 51 to 100 $396 13,031 10% Combined $ ,272 July 31, 2012 Page 72

76 Differences in administrative costs would also affect the premium impact seen under an expanded definition. Based on a study by The Lewin Group, administrative costs vary significantly based on the size of the group. The figure below indicates the results of this study. Group size is listed along the horizontal axis and administrative cost, as a percent of premium, is listed along the vertical axis. Figure 6: The Lewin Group Administrative Cost Study Source: The Lewin Group, Commonwealth Fund Commission on a High Performance Health System Most carriers provided information on the administrative cost load by submarket. The relativity of the small group admin cost to the 51 to 100 admin cost varied by carrier. Most small group admin costs were within plus or minus 5% of the 51 to 100 admin cost, with the exception of one outlier carrier where the small group admin cost was 12% lower than the 51 to 100 admin cost. The <51 Association and 51 to 100 Association admin costs were the same for all carriers, and the association admin costs were typically higher than the non-association admin costs. Noting the variance discussed above, the average admin cost for the <51 block was 1% lower than the admin cost for the 51 to 100 block (both including association business). This is contrary to the Lewin results discussed above. Given the 1% difference, on average, the admin cost should not have a large impact on the premium under an expanded small group definition, but could have an impact at the carrier level, given the fluctuation in relative admin costs by carrier. July 31, 2012 Page 73