563172
research-article2014
MCRXXX10.1177/1077558714563172Medical Care Research and ReviewMcCue and Hall
Data and Trends
Health Insurers’ Financial Performance and Quality Improvement Expenditures in the Affordable Care Act’s Second Year
Medical Care Research and Review 2015, Vol. 72(1) 113–122 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1077558714563172 mcr.sagepub.com
Michael J. McCue, DBA1 and Mark Hall, JD2
Abstract The Affordable Care Act requires health insurers to rebate any amounts less than 80%–85% of their premiums that they fail to spend on medical claims or quality improvement. This study uses the new comprehensive reporting under this law to examine changes in insurers’ financial performance and differences in their quality improvement expenditures. In the ACA’s second year (2012), insurers’ median medical loss ratios continued to increase and their median administrative cost ratios dropped, producing moderate operating margins in the group markets but a small operating loss in the individual market, at the median. For-profit insurers showed larger changes, in general, than did nonprofits. For quality improvement, insurers reported spending a significantly greater amount per member in their government plans than they did on their self-insured members, with spending on commercial insurance being in between these two extremes. The magnitude and source of these differences varied by corporate ownership. Keywords health insurers, financial performance, ACA
This article, submitted to Medical Care Research and Review on April 24, 2014, was revised and accepted for publication on October 24, 2014. 1Virginia 2Wake
Commonwealth University, Richmond, VA, USA Forest University, Winston-Salem, NC, USA
Corresponding Author: Michael J. McCue, DBA, Professor, Department of Health Administration, Virginia Commonwealth University, P. O. Box 980203, Richmond, VA 23298-0203, USA. Email: [email protected]
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
114
Medical Care Research and Review 72(1)
Background and New Contribution The Patient Protection and Affordable Care Act (ACA or “Affordable Care Act”) regulates health insurers’ medical loss ratios (MLRs) by requiring them to rebate to consumers any amounts less than 80%–85% of their premiums that they fail to spend on medical claims or quality improvement. This new federal regulation provides an unprecedented opportunity to observe various aspects of insurers’ financial performance that are important to public policy. Prior to this new law, insurers’ financial reporting, done through the National Association of Insurance Commissioners (NAIC), was geared mainly to state regulatory concerns and failed to capture some portions of the market, such as managed care plans in California that are not regulated by the state Department of Insurance. Since 2011, however, the ACA requires all health insurers to report comprehensive and somewhat more detailed financial information, in a uniform format, to the Centers for Medicare and Medicaid Services (CMS). This study takes advantage of this new data source to observe how the financial performance of health insurers’ commercial insurance is changing in response to new market regulations and structures. The ACA has initiated many changes in the health insurance market, and so it is difficult to predict what effects any single change, in isolation, will have on insurers’ financial performance. This study, however, focuses on data from 2011 to 2012—prior to the 2014 start date of the ACA’s major insurance market regulations. During this study’s time frame, the ACA’s most significant insurance regulation was of the MLR, from which we might expect several different responses. Insurers might reduce their administrative expenses and profits in order to avoid having to pay rebates. Instead, they might attempt to maximize profits by charging rates that are unlikely to exceed the mandated medical loss minimums and planning simply to pay rebates if their MLRs fall below the minimums. Moreover, the fact that quality improvement expenditures count as medical costs might encourage insurers to spend more on quality improvement. However, if that spending does not make their products more competitive or of greater value, they could decide instead to keep quality spending low in order to offer lower prices. Finally, insurers can be expected to adopt different strategies in different market segments and based on insurers’ corporate form. The ACA’s minimum MLRs were set at levels that were comfortably similar to market-wide averages for the group markets but that were higher than those prevailing in many individual markets. Therefore, more adaptation might be expected for the individual market, and some element of cross-subsidization could be possible between market segments. Also, for-profit insurers, which respond to investors’ profit expectations, can be expected to respond somewhat differently than nonprofits, which may tend to be more mission-oriented or less profit-driven. To explore these various possibilities, we assess changes from 2011 to 2012 in the MLR, administrative cost ratio, and profit margin ratio for commercial insurance, for nonprofit and for-profit insurers in each major market segment (individual, small group, and large group), and we measure differences in quality improvement expenditures for three different lines of insurance: commercial, government plans, and
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
115
McCue and Hall
self-insured employers. This study updates similar analyses in prior years (McCue, Hall and Liu, 2013; Abraham & Karaca-Mandic 2011) to see whether initial trends have continued and to gauge the industry’s evolving responses to the major regulatory and market changes under the Affordable Care Act.
Data and Methods To comply with the ACA’s regulation of MLRs, health insurers are required to file a reporting form with CMS’s Center for Consumer Information & Insurance Oversight (CCIIO) that provides key financial and membership data. These data files are reported at the CCIIO website as public use files.1 Multistate insurers are required to complete this form for each state in which they offer a comprehensive commercial health insurance business for any of the three market segments. We classified each insurer’s corporate ownership status (nonprofit vs. for-profit) based on information from their CMS reporting form, the Atlantic Information Services Directory of Health Plans, or the health insurer’s website. To compute the three main financial performance ratios (medical loss, administrative cost, and profit margin), financial data from the CMS forms are compared with the key summary information from insurers’ NAIC Supplemental Health Care Exhibit. Specifically, net premium begins with health premiums earned and then adjusts for high risk pools, reinsurance, taxes and assessments, and regulatory licenses and fees. This net premium serves as the denominator for each of the three key financial ratios. The MLR reflects total medical claims plus expenses incurred for activities to improve quality of care. The administrative cost ratio is based on total general administrative expenses and claims adjustment expenses. And the profit margin ratio reflects the insurer’s “underwriting” gain or loss—that is, the profit/loss from its insurance business, but not including any results from its financial investments. Also included in this study are detailed data regarding expenses that insurers incur to improve health care quality. In reporting these expenses, insurers allocate them across the following five quality improvement activities: those that improve health outcomes, prevent hospital readmission, improve patient safety/reduce medical errors, promote wellness and health, and support heath information technology to improve health care quality. To verify that insurers do not double-count such expenses across different products, some of which are less regulated, the government requires insurers to report quality improvement expenses for all of their lines of business, including self-insured plans that are not underwritten.2 Based on each health insurer’s reporting of membership in each state and market segment, in 2012 we identified 2,052 state health insurers offering individual insurance, 993 offering small group insurance, and 901 offering large group insurance. These numbers are marginally lower than in 2011, when 2,215 health insurers offered individual insurance, 1,058 offered small group insurance, and 934 offered large group insurance. We adjusted the initial sample for 2012 by applying the following exclusion criteria. First, we excluded health insurers with negative or zero values for net premium,
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
116
Medical Care Research and Review 72(1)
medical claims, or administrative costs; those with membership less than one; and 256 insurers that reported no financial data for 2011 (presumably because they were new to the market in 2012). Across all three market segments, the excluded insurers had substantially smaller membership.3 Second, for the individual market only, we adjusted the sample for extreme MLR values by excluding 25 health insurers with MLR values below the first percentile and above the 99th percentile.4 This selection produced the following final sample of state health insurers by market: 1,197 individual, 871 small group, and 737 large group. The distribution of insurers’ key financial performance ratios (unweighted) was highly variable and not normally shaped; therefore, we applied the nonparametric approach of the Wilcoxon signrank test. We first analyzed the commercial product data relative to the government and self-insured products by comparing their median expense per member values for the quality improvement activities as well as by corporate ownership.5 To analyze quality improvement expenses, our sample was somewhat different. Because insurers began to report these expenses only in 2011, we opted not to compare 2012 to 2011 because insurers may not have had systems fully in place the first year to capture and allocate these expenses. Also, because insurers report these expenses for lines of business other than their commercial products, such as government and selfinsured plans, we focus our analysis on differences across these lines of business rather than market segments within only their commercial lines. To improve actuarial credibility and avoid skewing by extreme outliers, we limited our quality expense analysis to insurers that reported values greater than zero and that had at least 1,000 members in the particular state and line of business. Given this sampling process, the study identified 795 health insurers offering a credible commercial plan and reporting quality care expenses data in 2012, 464 insurers that sold government plans, and 226 that administered self-insured products.6 We compared the median quality improvement expense per member for various types of insurers, product lines, and categories of quality expenses.
Study Results Key Financial Ratios For the majority of financial ratios in various market segments, the differences in median values were significant between 2011 and 2012 (see Table 1). The median MLR increased by 1.2 to 2.5 percentage points across different market segments. To offset this rise in the MLR, health insurers showed a significant reduction in their median administrative cost ratio in 2012, ranging from 0.7 to 1.9 percentage points in each market segment. Less pronounced was the change in median profit margins. This was significant only in the small group market, where the median increased by only 0.1 percentage points. Small profit margin declines in the individual and large group markets were not statistically significant.
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
117
McCue and Hall Table 1. Median Financial Ratios of All Insurers by Markets, 2011 and 2012. Financial Ratios
2011 (%)
Individual market Medical loss ratio Administrative cost ratio Profit margin Small group market Medical loss ratio Administrative cost ratio Profit margin Large group market Medical loss ratio Administrative cost ratio Profit margin
All n = 1,197 80.6% 18.9% –0.1% All n = 870 81.1% 15.3% 3.0% All n = 737 86.8% 11.0% 2.2%
2012 (%)
Changes from 2011 to 2012
83.1% 17.0% –1.1%
2.5%** –1.9%** –1.0%
82.3% 13.9% 3.1%
1.2%** –1.4%** 0.1%*
88.4% 10.3% 1.6%
1.6%** –0.7%** –0.6%
Source: CMS CCIIO MLR database for 2011 and 2012. *p < .05. **p < .01.
When analyzing these financial ratios by ownership category (see Table 2), we note that changes in these financial ratios were not significant among nonprofit health insurers, for whom the sample sizes ranged from 168 to 228 in each market segment. These changes were significant among for-profit insurers, however, which have larger sample sizes (509 to 1,029). Although we did not directly test for statistical significance based on ownership, it is interesting to note descriptively the pattern of changes in the financial ratios between for-profit and nonprofit insurers. MLRs increased similarly for both types of insurers, but for-profit insurers had a somewhat greater decline in their median administrative cost ratios. The range in decline in median for-profit administrative costs ratio was from 1.3 to 2.8 percentage points, compared with very little change among nonprofit insurers, except in the small group market, where their median administrative cost ratio dropped 0.7 percentage points. Accordingly, forprofit insurers had more favorable profit margins across all three market segments, with a smaller decline than nonprofits in the individual market, and a small increase, rather than decline, of median profit ratios in the small group market.
Quality Improvement Expenses For quality improvement expenses, Table 3 presents the median values across product types and by corporate ownership. To place these numbers in perspective, quality improvement expenses constitute less than 1% of premiums (0.88 %) for commercial insurers. Although modest in magnitude, relative differences between ownership types and product lines were significant for overall quality expense and for most subcategories except as otherwise noted in Table 3.
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
118
Medical Care Research and Review 72(1)
Table 2. Median Financial Ratios of Nonprofit vs For-Profit Insurers by Markets, 2011 and 2012. 2011 (%)
Financial Ratios
2012 (%)
Changes from 2011 to 2012
Individual market Nonprofit n = 168 Medical loss ratio 89.2% 90.9% Administrative cost ratio 13.6% 13.7% Profit margin –3.6% –4.3% Small group market Nonprofit n=219 Medical loss ratio 85.7% 86.9% Administrative cost ratio 11.6% 10.9% Profit margin 2.3% 1.5% Large group market Nonprofit n=228 Medical loss ratio 90.3% 90.9% Administrative cost ratio 8.8% 8.9% Profit margin 1.0% 0.6%
1.7% 0.1% –0.7% 1.2% –0.7%** –0.8% 0.6% 0.1% –0.4%
2011 (%)
2012 (%)
Changes from 2011 to 2012
For-Profit n = 1,029 79.0% 81.1% 2.1%* 20.2% 17.7% –2.5%* 0.4% 0.2% –0.2% For-Profit n = 651 79.1% 80.4% 1.3%** 18.2% 15.4% –2.8%** 3.4% 3.8% 0.4%* For-Profit n=509 84.8% 86.2% 1.4%** 12.5% 11.2% –1.3%** 3.1% 2.9% –0.2%
Source: CMS CCIIO MLR database for 2011 and 2012 and Atlantic Information Services 2012 Directory of Health Plans. *p < .05. **p < .01.
Table 3. Median Quality Improvement Expenses per Member by Ownership and by Product Line, 2012. Commercial Expense Total quality expense Health outcomes Wellness promotion Patient safety Readmission HIT
All
Government
For-profit Nonprofit
All
Self-Insured
For-profit Nonprofit
All
For-profit Nonprofit
n = 795
n = 578
n = 217
n = 464
n = 353
n = 111
n = 226
n = 159
n = 67
$30.1
$25.2
$35.2
$39.8
$28.1
$61.3
$12.5
$10.2
$26.7
$10.0
$8.3
$15.1
$18.6
$15.9
$29.3
$3.7
$1.7
$10.5
$2.4
$1.8
$4.5
$3.2
$2.3a
$4.9a
$5.5
$5.5
$2.8
$2.4 $1.8 $4.0b
$2.4a $1.7a $3.9
$2.3a $2.1a $4.8
$2.3 $2.5 $6.1
$1.9 $1.7 $6.5a
$4.0 $4.5 $5.6a
$0.4 $0.5 $2.1b
$0.1 $0.5 $2.1
$2.1 $2.0 $4.8
Note: All differences between corporate types and among product lines were significant, unless otherwise noted: a. not significant for corporate type within product line. b. not significant by product line (Commercial vs. Government and Commercial vs. Self-Insured). Source: CMS CCIIO MLR database for 2012.
Across all three product types (commercial, government, and self-insured), nonprofits spent more on quality improvement expenses per member than for-profit insurers. The greatest difference occurred for government and self-insured products, where nonprofits spent more than two times for-profits. Also across all three markets,
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
119
McCue and Hall
nonprofits allocated the majority of their dollars to health outcomes activities, but for-profit insurers did not do so for their self-insured members. In both the government and self-insured markets, nonprofits spent significantly more than for-profits on patient safety and preventing readmission activities. However within government products, for-profits spent significantly more on health information technology than did nonprofits. Comparing spending by type of product, health insurers reported spending 25% less per member per year (pmpy) for their commercial insurance products than for their government plans ($30 vs. $40 pmpy). This difference appears to be driven mainly by nonprofit insurers, who spent twice as much on quality improvement for their government plans ($61 pmpy) as did for-profit insurers ($28 pmpy). The greatest differential is seen for health outcomes activities, where both for-profit and nonprofit insurers spent about twice as much per member on their government plans. In contrast, insurers reported spending significantly less on quality improvement for their self-insured members. Overall, quality expenses per self-insured member were 58% lower than for commercial members and 69% lower than for government plans. These differences were greater among for-profit than nonprofit insurers. Nonprofit and for-profit insurers also reduced their quality improvement spending in different ways. Among for-profit insurers, reduced spending for self-insured members was most pronounced for expenses to improve health outcomes, on which for-profits spent $1.70 pmpy for self-insured members whereas nonprofit insurers spent $10.50 pmpy for self-insured members. In sharp contrast, for-profit insurers reported spending more on wellness promotion for their self-insured members ($5.50 pmpy), which is substantially higher than either for their commercial ($1.80) or government plan members ($2.3), and significantly higher than what nonprofits spent for wellness on their self-insured members ($2.80). For their commercial and government products, insurers allocated significantly more toward health outcomes activities than for the other quality improvement activities. Nonprofit insurers did the same for their self-insured members, but for-profit insurers allocated the most dollars toward wellness and health promotion activities for self-insured members.
Discussion Across all three market segments, federal regulation of MLRs appears to have prompted fully insured health insurers to become more efficient. Using data previously not available, we see that insurers’ median MLRs continue to increase and to exceed the regulated thresholds of 80% for individual and small group markets and 85% for the large group market in 2012. To counteract rising MLRs, health insurers in all three markets lowered their median administrative cost ratios, producing a median operating profit margin in the group markets of 3% and 1.6% for the small and large group markets, respectively.7 However, the reduction in administrative costs was not enough to avoid a median 1% operating loss in the individual market.
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
120
Medical Care Research and Review 72(1)
Nonprofit insurers earned a median profit of almost 1% on large groups and almost 1.5% on small groups, which could help offset any loss in the individual market. However, in the individual and small group markets, for-profit health insurers’ MLR was slightly above the 80% regulated threshold, compared with nonprofits MLRs of 91% and 87% (individual and small group). Thus, it appears that for-profit insurers may be targeting their MLR more closely to the regulated value in order to maintain a breakeven position in the individual market while keeping a more profitable position in the small group market. Prior work (McCue et al., 2013) reported similar patterns overall in 2011. However, differences emerged in 2012 based on corporate type. In 2011, the largest increases in the MLR and declines in profits occurred among for-profit insurers (whose median MLR increased by over 7 percentage points while their profit margin declined by 2 percentage points). In 2012, however, for-profit insurers’ median MLR increased by only 2 percentage points in the individual market while incurring no significant change in their profit margin. By reducing their administrative cost ratio, for-profit insurers were able to offset their rise in medical claims. In contrast, the median operating losses of nonprofit insurers in the individual market grew to 4% in 2012, driven by a rising median MLR of almost 91%. Substantial differences by corporate type and market segment were also observed for expenses to improve quality. Insurers spent the highest amount per member in their government plans and the lowest amount on their self-insured members. Data do not reveal why this disparity exists, but one possibility is that the Medicaid and Medicare programs require private insurers to report quality measures, which they actively measure. For self-insured products, because the clients tend to be large sophisticated employers, it is possible that employers take on more of the quality improvement functions themselves. That speculation is belied, however, by the fact that insurers usually use the same networks and care management programs for both insured and selfinsured members. Therefore, another possibility is that at least some insurers are using accounting conventions that allocate more of their shared quality improvement expenses to their regulated than to their nonregulated product lines. Another puzzle is why the government versus self-insured quality improvement disparity differs so distinctly between nonprofit and for-profit insurers. The greater quality improvement spending reported for government plans is driven mainly by nonprofit plans (median of $61 government vs. $35 commercial pmpy), whereas the lower spending on self-insured members is more pronounced among for-profit insurers ($10 self-insured vs. $25 commercial pmpy). Spending on health outcomes is one key differentiator, with nonprofits reporting about twice as much for this quality improvement category as for-profits for their commercial and government products, and six times more for their self-insured members. In contrast, for-profit plans report spending twice as much on wellness promotion for their self-insured members as do nonprofits. Further investigation is needed to determine whether these reports reflect substantive differences between these sectors or instead are based more on differing accounting allocation conventions.
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
121
McCue and Hall
Conclusion The ACA’s new regulation of MLRs has produced significant gains for consumers. In addition, by providing a larger window of insight into the internal finances of health insurers, it is a boon to the health services and health policy researchers. Peering through this window, in the second year of this regulation, we see that insurers continue to increase their median MLRs and reduce their median administrative cost ratios, leaving their operating profit margins with only small declines. We also see distinct differences between what health insurers spend on quality improvement activities among their different product lines (commercial, government, and self-insured), although explanations for these differences remain unclear. Finally, these patterns differ in intriguing ways between nonprofit and for-profit insurers. Further study is needed to determine why these patterns exist and whether they will persist. Declaration of Conflicting Interests The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors recognize the Commonwealth Fund for its financial support of this research.
Notes 1. The CCIIO website for MLR data is http://www.cms.gov/CCIIO/Resources/DataResources/mlr.html 2. Federal Register, December 1, 2010, p. 74875. 3. For those who reported no financial data in 2011, the median membership was 18 in the individual market, 220 in the small group market, and 205 in the large group market. Also, their 2012 financial ratios were not extraordinarily different than the overall market. Across the three market segments, their median MLRs ranged from 84% to 85%, and their median profit levels were –5.1% in the individual market, 1.7% small group, and 5.1% large group. 4. The MLR value at the 99th percentile was 367% and the first percentile value was 8.7%, compared with a median value overall of 83%. 5. Because we measured median rather than mean values for the various components of premiums and quality expenses, these component values do not sum to 100%. 6. These are independent but overlapping samples of insurers. Of the 795 commercial issuers, 399 offered only a commercial products with more than 1,000 members. Of the 464 government issuers, 112 offered only government plans with more than 1,000 members. Of the 226 self-insured carriers, 41 offered only self-insured products with more than 1,000 members. 7. Although we did not analyze the cost drivers of these administrative expenses, prior work (Herbold 2012) found that distribution costs—specifically direct salaries and commissions for agents and brokers—account for the highest percentage of administrative expenses within these markets.
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
122
Medical Care Research and Review 72(1)
References Abraham, J. M., & Karaca-Mandic, P. (2011). Regulating the medical loss ratio: Implications for the individual market. American Journal of Managed Care, 17, 211-218. Retrieved from http://www.ahipcoverage.com/wp-content/uploads/2011/03/AJMC-MLR-Paper.pdf Herbold, J. S. (2012). Administrative expenses: 2010 commercial health insurance. Milliman Research Report, February. Retrieved from http://publications.milliman.com/publications/ health-published/pdfs/commercial-health-insurance-admin-2010.pdf McCue, M., Hall, M., & Liu, X. (2013). The impact of the medical loss regulation on the financial performance of health insurers. Health Affairs, 32, 1546-1551.
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
research-article2014
MCRXXX10.1177/1077558714563172Medical Care Research and ReviewMcCue and Hall
Data and Trends
Health Insurers’ Financial Performance and Quality Improvement Expenditures in the Affordable Care Act’s Second Year
Medical Care Research and Review 2015, Vol. 72(1) 113–122 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1077558714563172 mcr.sagepub.com
Michael J. McCue, DBA1 and Mark Hall, JD2
Abstract The Affordable Care Act requires health insurers to rebate any amounts less than 80%–85% of their premiums that they fail to spend on medical claims or quality improvement. This study uses the new comprehensive reporting under this law to examine changes in insurers’ financial performance and differences in their quality improvement expenditures. In the ACA’s second year (2012), insurers’ median medical loss ratios continued to increase and their median administrative cost ratios dropped, producing moderate operating margins in the group markets but a small operating loss in the individual market, at the median. For-profit insurers showed larger changes, in general, than did nonprofits. For quality improvement, insurers reported spending a significantly greater amount per member in their government plans than they did on their self-insured members, with spending on commercial insurance being in between these two extremes. The magnitude and source of these differences varied by corporate ownership. Keywords health insurers, financial performance, ACA
This article, submitted to Medical Care Research and Review on April 24, 2014, was revised and accepted for publication on October 24, 2014. 1Virginia 2Wake
Commonwealth University, Richmond, VA, USA Forest University, Winston-Salem, NC, USA
Corresponding Author: Michael J. McCue, DBA, Professor, Department of Health Administration, Virginia Commonwealth University, P. O. Box 980203, Richmond, VA 23298-0203, USA. Email: [email protected]
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
114
Medical Care Research and Review 72(1)
Background and New Contribution The Patient Protection and Affordable Care Act (ACA or “Affordable Care Act”) regulates health insurers’ medical loss ratios (MLRs) by requiring them to rebate to consumers any amounts less than 80%–85% of their premiums that they fail to spend on medical claims or quality improvement. This new federal regulation provides an unprecedented opportunity to observe various aspects of insurers’ financial performance that are important to public policy. Prior to this new law, insurers’ financial reporting, done through the National Association of Insurance Commissioners (NAIC), was geared mainly to state regulatory concerns and failed to capture some portions of the market, such as managed care plans in California that are not regulated by the state Department of Insurance. Since 2011, however, the ACA requires all health insurers to report comprehensive and somewhat more detailed financial information, in a uniform format, to the Centers for Medicare and Medicaid Services (CMS). This study takes advantage of this new data source to observe how the financial performance of health insurers’ commercial insurance is changing in response to new market regulations and structures. The ACA has initiated many changes in the health insurance market, and so it is difficult to predict what effects any single change, in isolation, will have on insurers’ financial performance. This study, however, focuses on data from 2011 to 2012—prior to the 2014 start date of the ACA’s major insurance market regulations. During this study’s time frame, the ACA’s most significant insurance regulation was of the MLR, from which we might expect several different responses. Insurers might reduce their administrative expenses and profits in order to avoid having to pay rebates. Instead, they might attempt to maximize profits by charging rates that are unlikely to exceed the mandated medical loss minimums and planning simply to pay rebates if their MLRs fall below the minimums. Moreover, the fact that quality improvement expenditures count as medical costs might encourage insurers to spend more on quality improvement. However, if that spending does not make their products more competitive or of greater value, they could decide instead to keep quality spending low in order to offer lower prices. Finally, insurers can be expected to adopt different strategies in different market segments and based on insurers’ corporate form. The ACA’s minimum MLRs were set at levels that were comfortably similar to market-wide averages for the group markets but that were higher than those prevailing in many individual markets. Therefore, more adaptation might be expected for the individual market, and some element of cross-subsidization could be possible between market segments. Also, for-profit insurers, which respond to investors’ profit expectations, can be expected to respond somewhat differently than nonprofits, which may tend to be more mission-oriented or less profit-driven. To explore these various possibilities, we assess changes from 2011 to 2012 in the MLR, administrative cost ratio, and profit margin ratio for commercial insurance, for nonprofit and for-profit insurers in each major market segment (individual, small group, and large group), and we measure differences in quality improvement expenditures for three different lines of insurance: commercial, government plans, and
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
115
McCue and Hall
self-insured employers. This study updates similar analyses in prior years (McCue, Hall and Liu, 2013; Abraham & Karaca-Mandic 2011) to see whether initial trends have continued and to gauge the industry’s evolving responses to the major regulatory and market changes under the Affordable Care Act.
Data and Methods To comply with the ACA’s regulation of MLRs, health insurers are required to file a reporting form with CMS’s Center for Consumer Information & Insurance Oversight (CCIIO) that provides key financial and membership data. These data files are reported at the CCIIO website as public use files.1 Multistate insurers are required to complete this form for each state in which they offer a comprehensive commercial health insurance business for any of the three market segments. We classified each insurer’s corporate ownership status (nonprofit vs. for-profit) based on information from their CMS reporting form, the Atlantic Information Services Directory of Health Plans, or the health insurer’s website. To compute the three main financial performance ratios (medical loss, administrative cost, and profit margin), financial data from the CMS forms are compared with the key summary information from insurers’ NAIC Supplemental Health Care Exhibit. Specifically, net premium begins with health premiums earned and then adjusts for high risk pools, reinsurance, taxes and assessments, and regulatory licenses and fees. This net premium serves as the denominator for each of the three key financial ratios. The MLR reflects total medical claims plus expenses incurred for activities to improve quality of care. The administrative cost ratio is based on total general administrative expenses and claims adjustment expenses. And the profit margin ratio reflects the insurer’s “underwriting” gain or loss—that is, the profit/loss from its insurance business, but not including any results from its financial investments. Also included in this study are detailed data regarding expenses that insurers incur to improve health care quality. In reporting these expenses, insurers allocate them across the following five quality improvement activities: those that improve health outcomes, prevent hospital readmission, improve patient safety/reduce medical errors, promote wellness and health, and support heath information technology to improve health care quality. To verify that insurers do not double-count such expenses across different products, some of which are less regulated, the government requires insurers to report quality improvement expenses for all of their lines of business, including self-insured plans that are not underwritten.2 Based on each health insurer’s reporting of membership in each state and market segment, in 2012 we identified 2,052 state health insurers offering individual insurance, 993 offering small group insurance, and 901 offering large group insurance. These numbers are marginally lower than in 2011, when 2,215 health insurers offered individual insurance, 1,058 offered small group insurance, and 934 offered large group insurance. We adjusted the initial sample for 2012 by applying the following exclusion criteria. First, we excluded health insurers with negative or zero values for net premium,
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
116
Medical Care Research and Review 72(1)
medical claims, or administrative costs; those with membership less than one; and 256 insurers that reported no financial data for 2011 (presumably because they were new to the market in 2012). Across all three market segments, the excluded insurers had substantially smaller membership.3 Second, for the individual market only, we adjusted the sample for extreme MLR values by excluding 25 health insurers with MLR values below the first percentile and above the 99th percentile.4 This selection produced the following final sample of state health insurers by market: 1,197 individual, 871 small group, and 737 large group. The distribution of insurers’ key financial performance ratios (unweighted) was highly variable and not normally shaped; therefore, we applied the nonparametric approach of the Wilcoxon signrank test. We first analyzed the commercial product data relative to the government and self-insured products by comparing their median expense per member values for the quality improvement activities as well as by corporate ownership.5 To analyze quality improvement expenses, our sample was somewhat different. Because insurers began to report these expenses only in 2011, we opted not to compare 2012 to 2011 because insurers may not have had systems fully in place the first year to capture and allocate these expenses. Also, because insurers report these expenses for lines of business other than their commercial products, such as government and selfinsured plans, we focus our analysis on differences across these lines of business rather than market segments within only their commercial lines. To improve actuarial credibility and avoid skewing by extreme outliers, we limited our quality expense analysis to insurers that reported values greater than zero and that had at least 1,000 members in the particular state and line of business. Given this sampling process, the study identified 795 health insurers offering a credible commercial plan and reporting quality care expenses data in 2012, 464 insurers that sold government plans, and 226 that administered self-insured products.6 We compared the median quality improvement expense per member for various types of insurers, product lines, and categories of quality expenses.
Study Results Key Financial Ratios For the majority of financial ratios in various market segments, the differences in median values were significant between 2011 and 2012 (see Table 1). The median MLR increased by 1.2 to 2.5 percentage points across different market segments. To offset this rise in the MLR, health insurers showed a significant reduction in their median administrative cost ratio in 2012, ranging from 0.7 to 1.9 percentage points in each market segment. Less pronounced was the change in median profit margins. This was significant only in the small group market, where the median increased by only 0.1 percentage points. Small profit margin declines in the individual and large group markets were not statistically significant.
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
117
McCue and Hall Table 1. Median Financial Ratios of All Insurers by Markets, 2011 and 2012. Financial Ratios
2011 (%)
Individual market Medical loss ratio Administrative cost ratio Profit margin Small group market Medical loss ratio Administrative cost ratio Profit margin Large group market Medical loss ratio Administrative cost ratio Profit margin
All n = 1,197 80.6% 18.9% –0.1% All n = 870 81.1% 15.3% 3.0% All n = 737 86.8% 11.0% 2.2%
2012 (%)
Changes from 2011 to 2012
83.1% 17.0% –1.1%
2.5%** –1.9%** –1.0%
82.3% 13.9% 3.1%
1.2%** –1.4%** 0.1%*
88.4% 10.3% 1.6%
1.6%** –0.7%** –0.6%
Source: CMS CCIIO MLR database for 2011 and 2012. *p < .05. **p < .01.
When analyzing these financial ratios by ownership category (see Table 2), we note that changes in these financial ratios were not significant among nonprofit health insurers, for whom the sample sizes ranged from 168 to 228 in each market segment. These changes were significant among for-profit insurers, however, which have larger sample sizes (509 to 1,029). Although we did not directly test for statistical significance based on ownership, it is interesting to note descriptively the pattern of changes in the financial ratios between for-profit and nonprofit insurers. MLRs increased similarly for both types of insurers, but for-profit insurers had a somewhat greater decline in their median administrative cost ratios. The range in decline in median for-profit administrative costs ratio was from 1.3 to 2.8 percentage points, compared with very little change among nonprofit insurers, except in the small group market, where their median administrative cost ratio dropped 0.7 percentage points. Accordingly, forprofit insurers had more favorable profit margins across all three market segments, with a smaller decline than nonprofits in the individual market, and a small increase, rather than decline, of median profit ratios in the small group market.
Quality Improvement Expenses For quality improvement expenses, Table 3 presents the median values across product types and by corporate ownership. To place these numbers in perspective, quality improvement expenses constitute less than 1% of premiums (0.88 %) for commercial insurers. Although modest in magnitude, relative differences between ownership types and product lines were significant for overall quality expense and for most subcategories except as otherwise noted in Table 3.
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
118
Medical Care Research and Review 72(1)
Table 2. Median Financial Ratios of Nonprofit vs For-Profit Insurers by Markets, 2011 and 2012. 2011 (%)
Financial Ratios
2012 (%)
Changes from 2011 to 2012
Individual market Nonprofit n = 168 Medical loss ratio 89.2% 90.9% Administrative cost ratio 13.6% 13.7% Profit margin –3.6% –4.3% Small group market Nonprofit n=219 Medical loss ratio 85.7% 86.9% Administrative cost ratio 11.6% 10.9% Profit margin 2.3% 1.5% Large group market Nonprofit n=228 Medical loss ratio 90.3% 90.9% Administrative cost ratio 8.8% 8.9% Profit margin 1.0% 0.6%
1.7% 0.1% –0.7% 1.2% –0.7%** –0.8% 0.6% 0.1% –0.4%
2011 (%)
2012 (%)
Changes from 2011 to 2012
For-Profit n = 1,029 79.0% 81.1% 2.1%* 20.2% 17.7% –2.5%* 0.4% 0.2% –0.2% For-Profit n = 651 79.1% 80.4% 1.3%** 18.2% 15.4% –2.8%** 3.4% 3.8% 0.4%* For-Profit n=509 84.8% 86.2% 1.4%** 12.5% 11.2% –1.3%** 3.1% 2.9% –0.2%
Source: CMS CCIIO MLR database for 2011 and 2012 and Atlantic Information Services 2012 Directory of Health Plans. *p < .05. **p < .01.
Table 3. Median Quality Improvement Expenses per Member by Ownership and by Product Line, 2012. Commercial Expense Total quality expense Health outcomes Wellness promotion Patient safety Readmission HIT
All
Government
For-profit Nonprofit
All
Self-Insured
For-profit Nonprofit
All
For-profit Nonprofit
n = 795
n = 578
n = 217
n = 464
n = 353
n = 111
n = 226
n = 159
n = 67
$30.1
$25.2
$35.2
$39.8
$28.1
$61.3
$12.5
$10.2
$26.7
$10.0
$8.3
$15.1
$18.6
$15.9
$29.3
$3.7
$1.7
$10.5
$2.4
$1.8
$4.5
$3.2
$2.3a
$4.9a
$5.5
$5.5
$2.8
$2.4 $1.8 $4.0b
$2.4a $1.7a $3.9
$2.3a $2.1a $4.8
$2.3 $2.5 $6.1
$1.9 $1.7 $6.5a
$4.0 $4.5 $5.6a
$0.4 $0.5 $2.1b
$0.1 $0.5 $2.1
$2.1 $2.0 $4.8
Note: All differences between corporate types and among product lines were significant, unless otherwise noted: a. not significant for corporate type within product line. b. not significant by product line (Commercial vs. Government and Commercial vs. Self-Insured). Source: CMS CCIIO MLR database for 2012.
Across all three product types (commercial, government, and self-insured), nonprofits spent more on quality improvement expenses per member than for-profit insurers. The greatest difference occurred for government and self-insured products, where nonprofits spent more than two times for-profits. Also across all three markets,
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
119
McCue and Hall
nonprofits allocated the majority of their dollars to health outcomes activities, but for-profit insurers did not do so for their self-insured members. In both the government and self-insured markets, nonprofits spent significantly more than for-profits on patient safety and preventing readmission activities. However within government products, for-profits spent significantly more on health information technology than did nonprofits. Comparing spending by type of product, health insurers reported spending 25% less per member per year (pmpy) for their commercial insurance products than for their government plans ($30 vs. $40 pmpy). This difference appears to be driven mainly by nonprofit insurers, who spent twice as much on quality improvement for their government plans ($61 pmpy) as did for-profit insurers ($28 pmpy). The greatest differential is seen for health outcomes activities, where both for-profit and nonprofit insurers spent about twice as much per member on their government plans. In contrast, insurers reported spending significantly less on quality improvement for their self-insured members. Overall, quality expenses per self-insured member were 58% lower than for commercial members and 69% lower than for government plans. These differences were greater among for-profit than nonprofit insurers. Nonprofit and for-profit insurers also reduced their quality improvement spending in different ways. Among for-profit insurers, reduced spending for self-insured members was most pronounced for expenses to improve health outcomes, on which for-profits spent $1.70 pmpy for self-insured members whereas nonprofit insurers spent $10.50 pmpy for self-insured members. In sharp contrast, for-profit insurers reported spending more on wellness promotion for their self-insured members ($5.50 pmpy), which is substantially higher than either for their commercial ($1.80) or government plan members ($2.3), and significantly higher than what nonprofits spent for wellness on their self-insured members ($2.80). For their commercial and government products, insurers allocated significantly more toward health outcomes activities than for the other quality improvement activities. Nonprofit insurers did the same for their self-insured members, but for-profit insurers allocated the most dollars toward wellness and health promotion activities for self-insured members.
Discussion Across all three market segments, federal regulation of MLRs appears to have prompted fully insured health insurers to become more efficient. Using data previously not available, we see that insurers’ median MLRs continue to increase and to exceed the regulated thresholds of 80% for individual and small group markets and 85% for the large group market in 2012. To counteract rising MLRs, health insurers in all three markets lowered their median administrative cost ratios, producing a median operating profit margin in the group markets of 3% and 1.6% for the small and large group markets, respectively.7 However, the reduction in administrative costs was not enough to avoid a median 1% operating loss in the individual market.
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
120
Medical Care Research and Review 72(1)
Nonprofit insurers earned a median profit of almost 1% on large groups and almost 1.5% on small groups, which could help offset any loss in the individual market. However, in the individual and small group markets, for-profit health insurers’ MLR was slightly above the 80% regulated threshold, compared with nonprofits MLRs of 91% and 87% (individual and small group). Thus, it appears that for-profit insurers may be targeting their MLR more closely to the regulated value in order to maintain a breakeven position in the individual market while keeping a more profitable position in the small group market. Prior work (McCue et al., 2013) reported similar patterns overall in 2011. However, differences emerged in 2012 based on corporate type. In 2011, the largest increases in the MLR and declines in profits occurred among for-profit insurers (whose median MLR increased by over 7 percentage points while their profit margin declined by 2 percentage points). In 2012, however, for-profit insurers’ median MLR increased by only 2 percentage points in the individual market while incurring no significant change in their profit margin. By reducing their administrative cost ratio, for-profit insurers were able to offset their rise in medical claims. In contrast, the median operating losses of nonprofit insurers in the individual market grew to 4% in 2012, driven by a rising median MLR of almost 91%. Substantial differences by corporate type and market segment were also observed for expenses to improve quality. Insurers spent the highest amount per member in their government plans and the lowest amount on their self-insured members. Data do not reveal why this disparity exists, but one possibility is that the Medicaid and Medicare programs require private insurers to report quality measures, which they actively measure. For self-insured products, because the clients tend to be large sophisticated employers, it is possible that employers take on more of the quality improvement functions themselves. That speculation is belied, however, by the fact that insurers usually use the same networks and care management programs for both insured and selfinsured members. Therefore, another possibility is that at least some insurers are using accounting conventions that allocate more of their shared quality improvement expenses to their regulated than to their nonregulated product lines. Another puzzle is why the government versus self-insured quality improvement disparity differs so distinctly between nonprofit and for-profit insurers. The greater quality improvement spending reported for government plans is driven mainly by nonprofit plans (median of $61 government vs. $35 commercial pmpy), whereas the lower spending on self-insured members is more pronounced among for-profit insurers ($10 self-insured vs. $25 commercial pmpy). Spending on health outcomes is one key differentiator, with nonprofits reporting about twice as much for this quality improvement category as for-profits for their commercial and government products, and six times more for their self-insured members. In contrast, for-profit plans report spending twice as much on wellness promotion for their self-insured members as do nonprofits. Further investigation is needed to determine whether these reports reflect substantive differences between these sectors or instead are based more on differing accounting allocation conventions.
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
121
McCue and Hall
Conclusion The ACA’s new regulation of MLRs has produced significant gains for consumers. In addition, by providing a larger window of insight into the internal finances of health insurers, it is a boon to the health services and health policy researchers. Peering through this window, in the second year of this regulation, we see that insurers continue to increase their median MLRs and reduce their median administrative cost ratios, leaving their operating profit margins with only small declines. We also see distinct differences between what health insurers spend on quality improvement activities among their different product lines (commercial, government, and self-insured), although explanations for these differences remain unclear. Finally, these patterns differ in intriguing ways between nonprofit and for-profit insurers. Further study is needed to determine why these patterns exist and whether they will persist. Declaration of Conflicting Interests The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors recognize the Commonwealth Fund for its financial support of this research.
Notes 1. The CCIIO website for MLR data is http://www.cms.gov/CCIIO/Resources/DataResources/mlr.html 2. Federal Register, December 1, 2010, p. 74875. 3. For those who reported no financial data in 2011, the median membership was 18 in the individual market, 220 in the small group market, and 205 in the large group market. Also, their 2012 financial ratios were not extraordinarily different than the overall market. Across the three market segments, their median MLRs ranged from 84% to 85%, and their median profit levels were –5.1% in the individual market, 1.7% small group, and 5.1% large group. 4. The MLR value at the 99th percentile was 367% and the first percentile value was 8.7%, compared with a median value overall of 83%. 5. Because we measured median rather than mean values for the various components of premiums and quality expenses, these component values do not sum to 100%. 6. These are independent but overlapping samples of insurers. Of the 795 commercial issuers, 399 offered only a commercial products with more than 1,000 members. Of the 464 government issuers, 112 offered only government plans with more than 1,000 members. Of the 226 self-insured carriers, 41 offered only self-insured products with more than 1,000 members. 7. Although we did not analyze the cost drivers of these administrative expenses, prior work (Herbold 2012) found that distribution costs—specifically direct salaries and commissions for agents and brokers—account for the highest percentage of administrative expenses within these markets.
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
122
Medical Care Research and Review 72(1)
References Abraham, J. M., & Karaca-Mandic, P. (2011). Regulating the medical loss ratio: Implications for the individual market. American Journal of Managed Care, 17, 211-218. Retrieved from http://www.ahipcoverage.com/wp-content/uploads/2011/03/AJMC-MLR-Paper.pdf Herbold, J. S. (2012). Administrative expenses: 2010 commercial health insurance. Milliman Research Report, February. Retrieved from http://publications.milliman.com/publications/ health-published/pdfs/commercial-health-insurance-admin-2010.pdf McCue, M., Hall, M., & Liu, X. (2013). The impact of the medical loss regulation on the financial performance of health insurers. Health Affairs, 32, 1546-1551.
Downloaded from mcr.sagepub.com at CARLETON UNIV on May 9, 2015
