New Mexico Health Connections v. United States Department of Health and Human Services et al

Filing 55

MEMORANDUM OPINION AND ORDER by District Judge James O. Browning granting in part and denying in part the Plaintiff's Motion for Summary Judgment 32 and the Defendants' Cross-Motion for Summary Judgment 34 ; and, denying the Defendants' Motion for Leave to File Reply Concerning Notice of Supplemental Authority 51 . (kw)

IN THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF NEW MEXICO
NEW MEXICO HEALTH CONNECTIONS,
a New Mexico Non-Profit Corporation,
Plaintiff,
vs.

No. CIV 16-0878 JB/JHR

UNITED STATES DEPARTMENT OF
HEALTH AND HUMAN SERVICES;
CENTERS FOR MEDICARE AND
MEDICAID SERVICES; SYLVIA MATHEWS
BURWELL, Secretary of the United States
Department of Health and Human Services, in
her official capacity and ANDREW M.
SLAVITT, Acting Administrator for the Centers
for Medicare and Medicaid Services, in his
official capacity,
Defendants.
MEMORANDUM OPINION AND ORDER
THIS MATTER comes before the Court on: (i) the Plaintiff’s Motion for Summary
Judgment, filed April 13, 2017 (Doc. 32)(“Health Connection’s Motion”); and (ii) the
Defendants’ Cross-Motion for Summary Judgment, filed June 1, 2017 (Doc. 34)(“Defendants’
Motion”). The Court held a hearing on January 22, 2018. The primary issues are: (i) whether
the Administrative Procedure Act, 5 U.S.C. § 702 (“APA”), waives sovereign immunity for all
of Plaintiff New Mexico Health Connections’ claims; (ii) whether incorporating statewide
average premiums in Defendant United States Department of Health and Human Services’
(“HHS”) 1 risk-adjustment formula is contrary to law or arbitrary and capricious; (iii) whether
HHS’ approach to predicting costs for hierarchal condition category (“HCC”) and non-HCC
1

HHS is not the only Defendant in this case, but Health Connections challenges the
agency’s actions, so the Court will refer to HHS only, for simplicity’s sake.

eligible enrollees is arbitrary and capricious; (iv) whether HHS’ decisions regarding partial year
enrollees and the use of prescription drug data in its risk adjustment model are arbitrary and
capricious; and (v) whether HHS’ risk adjustment formula effectively bans bronze health
insurance plans and is contrary to law. The Court concludes that: (i) the APA waives sovereign
immunity for all of the claims presented, thereby giving the Court subject-matter jurisdiction;
(ii) HHS’ use of statewide average premiums in its risk adjustment methodology is not contrary
to law, but is arbitrary and capricious; (iii) HHS’ approach to predicting costs for HCC and nonHCC eligible enrollees is not arbitrary and capricious; (iv) HHS’ decisions regarding partial year
enrollees and the use of prescription drug data in its risk adjustment model are not arbitrary and
capricious; and (v) HHS’ risk adjustment formula does not, in effect, ban bronze health insurance
plans. Accordingly, the Health Connection’s Motion is granted in part and denied in part. The
Defendants’ Motion is granted in part and denied in part. The Court sets aside and vacates the
agency action as to the statewide average premium rules and remands the case to the agency for
further proceedings.

It otherwise dismisses Health Connections’ remaining claims with

prejudice.
FACTUAL BACKGROUND
Health Connections seeks APA review of agency action, so rule 56 of the Federal Rules
of Civil Procedure does not apply even though both Health Connections and HHS ostensibly
filed motions for summary judgment. See Olenhouse v. Commodity Credit Corp., 42 F.3d 1560,
1580 (10th Cir. 1994)(“Reviews of agency action in the district courts must be processed as
appeals.” (emphasis in original))(“Olenhouse”); id. (“[M]otions for summary judgment are
conceptually incompatible with the very nature and purpose of an appeal.”). See also Jarita
Mesa Livestock Grazing Ass’n v. U.S. Forest Serv., 305 F.R.D. 256, 281 (D.N.M.

-2-

2015)(Browning, J.). Accordingly, district courts reviewing agency action do not determine
whether a “genuine dispute as to any material fact” exists, Fed. R. Civ. P. 56, and instead
“engage in a substantive review of the record to determine if the agency considered relevant
factors or articulated a reasoned basis for its conclusions,” Olenhouse, 42 F.3d at 1580. See
Jarita Mesa Livestock Grazing Ass’n v. U.S. Forest Serv., 305 F.R.D at 281 (“District courts
may not entertain motions for summary judgment or any other procedural devices that shift the
appellant’s substantial burden -- arbitrary-or-capricious review for questions of fact and Chevron
deference for questions of statutory interpretation -- onto the agency.”). While engaging in that
substantive review, “the district court should govern itself by referring to the Federal Rules of
Appellate Procedure.”

Olenhouse, 42 F.3d at 1580.

To be clear, the Court recounts the

following undisputed facts as a comprehensive factual background for its APA review and not as
a summary-judgment analysis.2

2

The Court’s approach when reviewing agency action is importantly different from its
summary-judgment approach, because “judicial review of agency action is normally restricted to
the administrative record.” Lee v. U.S. Air Force, 354 F.3d 1229, 1242 (10th Cir. 2004). See
Camp v. Pitts, 411 U.S. 138, 142 (1973)(“[T]he focal point for judicial review should be the
administrative record already in existence, not some new record made initially in the reviewing
court.”). The original administrative record before the Court, see Defendants’ Notice of Manual
Filing of Administrative Record, filed February 22, 2017 (Doc. 25), inadvertently omits some
documents, however, see Stipulation Concerning Administrative Record ¶ 1, at 1, filed March
15, 2017 (Doc. 29), and “[i]f anything material to either party is omitted from or misstated in the
record by error or accident, the omission or misstatement may be corrected and a supplemental
record may be certified,” Fed. R. App. P. 10(e)(2). Health Connections filed a “Supplemental
Appendix of record materials,” Plaintiff New Mexico Health Connections’ Notice of Manual
Filing of Supplemental Appendix at 1, filed April 13, 2017 (Doc. 31). The Court will cite that
supplemental appendix as “NMHC.”
Health Connections argues that the Court should further supplement the administrative
record by taking judicial notice of certain documents. See Plaintiff’s Reply and Opposition to
Defendants’ Cross-Motion for Summary Judgment at 11 n.13, filed July 13, 2017 (Doc. 40).
Those documents include newspaper articles, state agency websites, federal agency reports, and
congressional reports. See Exhibit 5, filed April 13, 2017 (Doc. 33-2)(congressional report);
Exhibit 13, filed April 13, 2017 (Doc. 33-6); Exhibit 14, filed April 13, 2017 (Doc. 33-7)(state
agency website; Exhibit A, filed July 13 2017 (Doc. 40-2)(newspaper article).
-3-

1.

The Affordable Care Act.

Congress enacted The Patient Protection and Affordable Care Act, Pub. L. No. 111-148,
124 Stat. 119 (2010)(codified at 42 U.S.C. §§ 300gg-1 to -19, 18001-18022)(“ACA”) “to expand
coverage in the individual health insurance market.” King v. Burwell, 135 S. Ct. 2480, 2485
(2015)(Roberts, C.J.). To effect that goal, the ACA: (i) bars insurers from considering preexisting medical conditions when deciding whether to sell insurance and determining prices;
(ii) requires individuals to make an individual shared responsibility payment to the Internal

Rule 201 of the Federal Rules of Evidence permits judicial notice of facts that are “not
subject to reasonable dispute,” because those facts are either “generally known” or “can be
accurately and readily determined from sources whose accuracy cannot reasonably be
questioned.” Fed. R. Evid. 201(b). The United States Court of Appeals for the Tenth Circuit’s
precedent indicates that the ordinary evidentiary rules regarding judicial notice apply when a
court reviews agency action. See New Mexico ex. rel. Richardson v. Bureau of Land Mgmt.,
565 F.3d 683, 702 n.21 (10th Cir. 2009)(citing Fed. R. Evid. 201(b))(“We take judicial notice of
this document, which is included in the record before us in [another case].”); id. at 702 n.22
(“We conclude that the occurrence of Falcon releases is not subject to reasonable factual dispute
and is capable of determination using sources whose accuracy cannot reasonably be questioned,
and we take judicial notice thereof.”). In contrast, the United States Courts of Appeals for the
Ninth and Eleventh Circuits have held that taking judicial notice is inappropriate in APA reviews
absent extraordinary circumstances or inadvertent omission from the administrative record. See
Compassion Over Killing v. U.S. Food & Drug Administration, 849 F.3d 849, 852 n.1 (9th Cir.
2017); National Min. Ass’n v. Secretary U.S. Dep’t of Labor, 812 F.3d 843, 875 (11th Cir.
2016).
The Court will not use rule 201 of the Federal Rules of Evidence to add the documents
that Health Connections identifies to the administrative record, because rule 201 permits the
Court to take judicial notice of facts and not documents. That a document exists is a fact and,
with respect to the documents that Health Connections filed with the Court, it is a fact
susceptible to judicial notice. See Graham v. Catamaran Health Solutions, LLC, 2017 WL
3613328, at *2 n.1 (8th Cir. 2017)(“[W]e may take judicial notice of filings of public record and
the fact (but not the veracity) of parties’ assertions therein.”); In re Santa Fe Natural Tobacco
Company Marketing & Sales Practices and Products Litig., __ F. Supp. 3d __, 2017 WL
6550897, at *61 (D.N.M. 2017)(Browning, J.)(“Indeed, any party can file a document in a
proceeding, but that does not mean that the document’s contents are beyond reproach.”). It does
not follow from a document’s existence, however, that it contains facts that are “not subject to
reasonable dispute.” Fed. R. Evid. 201(b). See Stephen A. Saltzburg et al., FEDERAL RULES OF
EVIDENCE MANUAL § 201.02[3], at 201-08 (“[A] court can take judicial notice that court filings
contained certain allegations . . . , [b]ut the truth of these allegations and findings are not proper
subjects of judicial notice.”).
-4-

Revenue Service unless they maintain health-insurance coverage; and (iii) gives certain
individuals tax credits to make health insurance more affordable for them. See King v. Burwell,
135 S. Ct. at 2485; 26 U.S.C. § 5000A (describing the individual shared responsibility payment
requirement).
Additionally, the ACA establishes Health Benefit Exchanges (“Exchanges”), online
marketplaces where individuals can purchase health insurance and potentially obtain federal
subsidies. See 42 U.S.C. §§ 18031-18033. Qualified health plans sold on the Exchanges must
provide bronze-level, silver-level, gold-level, or platinum-level coverage.

See 42. U.S.C.

§ 18021(a)(1)(defining a qualified health plan); 42 U.S.C. § 18022(d)(1)(setting out four
coverage levels). Bronze-level plans are designed such that, on average, the insurance company
pays sixty percent of its policyholders’ covered healthcare costs; that percentage increases to
seventy, eighty, and ninety percent for silver-, gold-, and platinum-level plans, respectively. See
42 U.S.C. § 18022(d)(1); The ‘Metal’ Categories: Bronze, Silver, Gold & Platinum,
HEALTHCARE.GOV, http://www.healthcare.gov/choose-a-plan/plans-categories/.

Consequently,

bronze-level plans tend to attract individuals who anticipate fewer healthcare needs, i.e.,
healthier people, whereas gold-level and platinum-level plans tend to attract individuals who
anticipate more healthcare needs, i.e., sicker individuals. See State Health Insurance Exchange
Risk Adjustment and Plan Metals Level Memorandum at 3 (dated December 15,
2011)(A.R.000811); Risk Adjustment Implementation Issues, Draft for Discussion Purposes at
31 (dated September 12, 2011)(A.R.004397).
The ACA also establishes the Consumer Operated and Oriented Plan (“CO-OP”)
program.

42 U.S.C. § 18042(a).

The CO-OP program provides loans and grants to new

nonprofit health-insurance issuers, which fosters competition in the individual health-insurance

-5-

market. 42 U.S.C. § 18042(a)-(b). See also Memorandum of Law in Support New Mexico
Health Connections’ Motion For Summary Judgment ¶ 19, at 10, filed April 13, 2017
(Doc. 33)(“Plaintiff Mem.”)(“Congress created the CO-OP program to enhance competition.”).3
To receive these loans or grants, however, insurers must offer their health-insurance plans on the
Exchanges. See 45 C.F.R. § 156.515(c). See also Plaintiff Mem. ¶ 21, at 10.
The ACA expands healthcare access, but it also increases health-insurance-industry risk.
That the ACA requires insurers to cover all individuals, healthy or otherwise, means an unlucky
insurer could end up providing coverage to a particularly sickly group of customers. See 42
U.S.C. § 300gg-1(a)(“[E]ach health insurance issuer that offers health insurance coverage in the
individual or group market in a State must accept every employer and individual in the State that
applies for such coverage.”). The ACA makes things even worse for those unlucky insurers by
prohibiting them from responding to the increased cost of providing healthcare coverage to
sicker individuals by charging those individuals higher prices.

See 42 U.S.C.

§ 300gg(a)(prohibiting price discrimination based on factors other than geography, age, tobacco
use, and whether coverage extends to an individual to a family). Taken together, those two ACA
requirements “threaten to impose massive new costs on insurers, who are required to accept
unhealthy individuals but prohibited from charging them rates necessary to pay for their
coverage.” Nat’l Fed’n of Indep. Bus. v. Sebelius, 567 U.S. 519, 548 (2012).
The ACA contemplates three kinds of programs -- two temporary and one permanent -that ameliorate the risks it creates. See 42 U.S.C. §§ 18061-18063. First, under transitional
reinsurance programs, which operate only from 2014 to 2016, insurers make payments to “an
applicable reinsurance entity,” typically HHS, and reinsurance entities use those funds to provide
3

Health Connections’ arguments supporting its motion are all in the Plaintiff Mem. See
Plaintiff Mem. at 1.
-6-

“reinsurance payments” to insurers that cover “high risk individuals.” 42 U.S.C. § 18061(b)(1).
According to HHS, “[t]he reinsurance program will reduce the uncertainty of insurance risk in
the individual market by partially offsetting issuers’ risk associated with high-cost enrollees.”
HHS Notice of Benefit and Payment Parameters for 2014, 78 Fed. Reg. 15,410, 15,411 (dated
March 11, 2013)(A.R.000227-28)(“2014 Final Rule”). “Each State is eligible to establish a
reinsurance program,” but “HHS will establish a reinsurance program for each State that does
not elect to establish its own reinsurance program.” 45 C.F.R. § 153.210(a), (c). Second, under
the temporary risk corridor program, which also operates only from 2014 to 2016, sufficiently
profitable insurers must make payments to HHS while HHS must make payments to sufficiently
unprofitable insurers. See 42 U.S.C. § 18062. Those payments, HHS predicts, “will protect
against uncertainty in rate setting for qualified health plans by limiting the extent of issuers’
financial losses and gains.” 2014 Final Rule, 78 Fed. Reg. at 15,411 (A.R.000228).
Third, under permanent risk adjustment programs, “each State shall assess a charge” on
insurers “if the actuarial risk of [their] enrollees . . . for a year is less than the average actuarial
risk of all enrollees in all plans or coverage in such State for such year.”

42 U.S.C.

§ 18063(a)(1). Likewise, “each State shall provide a payment” to insurers “if the actuarial risk of
[their] enrollees . . . is greater than the average actuarial risk of all enrollees in all plans and
coverage in such State for such year.” 42 U.S.C. § 18063(a)(2). Risk adjustment programs are
“intended to provide increased payments to health insurance issuers that attract higher-risk
populations, such as those with chronic conditions, and reduce the incentives for issuers to avoid
higher-risk enrollees.” 2014 Final Rule, 78 Fed. Reg. at 15,411 (A.R.000228).
2.

Risk Adjustment Implementation.

While the ACA refers to “States” assessing charges and providing payments in risk

-7-

adjustment programs, 42 U.S.C. § 18063(a), it also tells HHS to, “in consultation with States,
shall establish criteria and methods to be used in carrying out the risk adjustment activities,” 42
U.S.C. § 18063(b), and HHS regulations state that it will implement PRA programs for “[a]ny
State that does not elect to operate an Exchange, or that HHS has not approved to operate an
Exchange,” 45 C.F.R. § 153.310(a)(2), for “[a]ny State that elects to operate an Exchange but
does not elect to administer risk adjustment,” 45 C.F.R. § 153.310(a)(3), and for, “[b]eginning in
2015, any State that is approved to operate an Exchange and elects to operate risk adjustment but
has not been approved by HHS to operate risk adjustment,” 45 C.F.R. § 153.310(a)(4). Only
Massachusetts, however, elected to operate its own PRA program, see HHS Notice of Benefit
and Payment Parameters for 2016, 80 Fed. Reg. 10,750, 10,759 (dated February 27, 2015)(A.R.
005691)(“2016 Final Rule”), and that program did not last long, see HHS Notice of Benefit and
Payment

Parameters

for

2017,

81

Fed.

Reg.

12,204,

12,230

(dated

March

8,

2016)(A.R.007774)(“2017 Final Rule”)(“We are not recertifying the alternate State methodology
for use in Massachusetts for 2017 risk adjustment. Massachusetts and HHS will begin the
transition that will allow HHS to operate risk adjustment in Massachusetts in 2017.”).
HHS thus implements New Mexico’s -- and forty-nine other states’ -- risk adjustment
program.

See

2017

Final

Rule,

81

Fed.

Reg.

at

12,230

(dated

March

8,

2016)(A.R.007774)(“HHS will operate risk adjustment in all States for the 2017 benefit year.”).
In doing so, HHS annually publishes its risk adjustment methodology. See 45 C.F.R. § 153.320
(“HHS will specify in the annual HHS notice of benefit and payment parameters for the
applicable year the Federally certified risk adjustment methodology that will apply in States that
do not operate a risk adjustment program.”). HHS’ published risk adjustment methodology must
describe: (i) how HHS calculates individual risk scores, see 45 C.F.R. § 153.320(b)(1), which are

-8-

“a relative measure of predicted health care costs” for particular individuals, 45 C.F.R. § 153.20;
(ii) how HHS determines a plan’s average actuarial risk from individual risk scores, see 45
C.F.R. §§ 153.20, .320(b)(2); and (iii) how HHS uses a plan’s average actuarial risk to determine
the plan’s risk adjustment payments and charges, see 45 CFR §§ 153.20, .320(b)(3).
HHS’ risk adjustment methodology4 “predict[s] plan liability for an enrollee based on
that person’s age, sex, and diagnoses (risk factors), producing a[n individual] risk score.” 2014
Final Rule, 78 Fed. Reg. at 15,419. HHS calculates a health plan’s average risk score by
averaging its enrollees’ individual risk scores; each individual risk score is weighted by the
number of months the relevant individual was enrolled in the health plan. See 2014 Final Rule,
78 Fed. Reg. at 15,432. HHS multiplies the “State average premium” by several plan-cost
factors, “relative measures that compare how [a] plan[] differ[s] from the market average,”
including the plan’s average risk score to produce to produce a plan-premium estimate. 2014
Final Rule, 78 Fed. Reg. at 15,430-31. “Multiplying the plan[’s] average risk score by the State
average premium shows how a plan’s premium would differ from the State average premium
based on the risk selection experienced by the plan.” 2014 Final Rule, 78 Fed. Reg. at 15,431.
HHS also produces a second plan-premium estimate by multiplying the state average premium
by plan-cost factors other than the plan’s average risk score. 2014 Final Rule, 78 Fed. Reg. at
15,430. HHS’ payment transfer formula takes the first plan-premium estimate and subtracts the
second, which “provides a per member per month (PMPM) transfer amount for a plan.” 2014
4

The Court refers to HHS’ published risk adjustment methodology in general terms even
though HHS has five different published risk adjustment methodologies, one for each year from
2014 to 2018, because, while those methodologies differ in detail, they have the same basic
structure. See, e.g., 81 Fed. Reg. 12,330 (dated March 8, 2016)(“Although we did not propose to
change the payment transfer formula from what was finalized in the 2014 Payment in its entirety,
since, as noted above, we are recalibrating the HHS risk adjustment model.”). Where the
differences between HHS’ five payment methodologies are important, the Court will be more
specific.

-9-

Final Rule, 78 Fed. Reg. at 15,431. Finally, HHS multiplies a plan’s per member, per month
transfer amount by its number of “billable member months . . . to calculate the plan’s total risk
adjustment payment.” 2014 Final Rule, 78 Fed. Reg. at 15,431.
Each year, HHS monitors and updates the risk adjustment model “with more recent data,”
but it does not “reconsider[] the entire methodology anew each year.” Defendant Mem. ¶ 14, at
13. See Defendant Mem. ¶ 15, at 13; Plaintiff Mem. ¶ 7, at 5-6. There is a lag between HHS’
promulgation of annual risk adjustment formula rules and the data it received from issuers, so,
“[b]y the time results for the program’s first year (2014) were announced,” HHS had already
promulgated its annual rules for 2015 and 2016. Defendant Mem. ¶¶ 4-6, 16, at 10, 14 (noting
that it takes two calendar years between publication of a benefit rule and the announcement of
risk adjustment payments under that rule). See Plaintiff Mem. ¶ 11, at 7. For its 2017 rule, HHS
updated its methodology for future years based on data it had collected from its 2014 results.
See Defendant Mem. ¶ 16, at 14 (citing 2017 Final Rule, 81 Fed. Reg. at 12,218-20
(A.R.007762-64)). HHS adjusted its 2018 rule to account for partial-year enrollees and began
using limited pharmaceutical data to help measure individuals’ relative healthiness. Defendant
Mem. ¶ 18, at 14 (citing HHS Notice of Benefit and Payment Parameters of 2018, 81 Fed. Reg.
94,058, 94,072-76 (dated December 22, 2016)(A.R.009609-13)(“2018 Final Rule”)). Health
Connections supported the partial-year enrollee adjustment, but urged HHS to apply the
adjustment retroactively to risk adjustment transfers for 2014 and 2015. See Declaration of
Martin Hickey, MD ¶ 98, at 24 (dated October 5, 2016)(NMHC000886)(“Hickey Declaration”).
3.

Health Connections.

Health Connections is a CO-OP program participant, and it has operated in New Mexico
since 2014. See Hickey Declaration ¶ 27, at 5 (NMHC000867). Health Connections signed a

- 10 -

loan agreement with HHS to fund Health Connections’ initial formation and New Mexico
operations. See Hickey Declaration ¶ 28, at 5 (NMHC000867). Health Connections began
enrolling members in October, 2013 and providing coverage in January, 2014. See Hickey
Declaration ¶ 27, at 5 (NMHC000867). Health Connections has grown from 14,000 members in
2014, to 44,500 members in 2016. See Hickey Declaration ¶ 33, at 6 (NMHC000868).
Health Connections offers -- and has offered since its inception -- the lowest or secondlowest cost health insurance plan in New Mexico.

See Hickey Declaration ¶ 31, at 6

(NMHC000868). It has offered such affordable plans even while serving unhealthy enrollees;
New Mexico has the highest prevalence of Hepatitis C in the nation, and Health Connections
enrolled many members of that population in its plans. See Patient Protection and Affordable
Care Act Comments to HHS Notice of Benefit and Payment Parameters for 2018 at 19-20 (dated
October 6, 2016)(NMHC0000853-54)(“2018 Comments”).

At a meeting of the National

Association of Insurance Commissioners, the Superintendent of Insurance of New Mexico stated
that Health Connections’ entry into the health-insurance marketplace increased competition and
saved New Mexicans over half a billion dollars.

See Hickey Declaration ¶ 36, at 7,

(NMHC000869).
While many health-insurance companies aim for a profit margin between two and five
percent of their premiums, see Hickey Declaration ¶ 19, at 4 (NMHC000866), for 2014, many
small health-insurance companies were required to pay over ten percent of their premiums as
risk-adjustment charges, see Centers for Medicare & Medicaid Services, United States
Department

of

Health

and

2016)(NMHC001018)(“Choices”).

Human

Services,

Choices

at

2

(dated

April

22,

For that year, HHS assessed Health Connections a

$6,666,798.00 risk-adjustment charge, which is equal to 21.5% of Health Connections’ 2014

- 11 -

premiums. See Hickey Declaration ¶ 17, at 3, (NMHC000865). For 2015, HHS assessed Health
Connections a $14,569,495.74 risk-adjustment charge, which is equal to 14.7% of Health
Connections’ 2015 premiums. See Hickey Declaration ¶ 18, at 4 (NMHC000866).
Risk-adjustment charges have, thus, forced several CO-OP program participants to close
their doors.

See 2018 Comments at 3 (NMHC000837); U.S. House of Representatives

Committee on Energy and Commerce, Implementing Obamacare: A Review of CMS’
Management

of

the

Failed

CO-OP

Program

at

19-22

(dated

September

13,

2016)(NMHC000910-13); Technical Issues with ACA Risk Adjustment and Risk Corridor
Programs, and Financial Impact on New, Fast-Growing, and Efficient Health Plans at 11-13,
(NMHC001000-02); Connecticut Insurance Department, Insurance Department Places
HealthyCT Under Order of Supervision, at 1-2 (dated September 26, 2016)(NMHC001351-52)).
Several state insurance commissioners have expressed concern about the risk adjustment
program. For example, Maryland’s Insurance Commissioner testified to Congress:
Over the past few years, new innovative health insurance plans have been created
that are providing enhanced competition and patient care. And it is working. For
year-end 2014, Carefirst had a 91% market share of the individual market in
Maryland. Today, it is 57%, due in part to a more competitive marketplace.
These carriers have the potential to continue but their ability to do so is severely
jeopardized by the adverse and perhaps fatal financial impact caused by the
technical shortcoming of the current risk adjustment and risk corridor programs.
....
The risk adjustment formula is of concern to state regulators because it has
proven to place newer carriers at a distinct disadvantage. For example, the risk
adjustment formula quantifies an enrollee’s health status based on age and
diagnoses recorded during the course of the year. New carriers have very limited
information on the health status or previous claims history of the applicants.
Therefore, the carrier’s population may appear healthier than it actually is if some
diagnoses are not captured which may result in improper risk adjustment
payments.
See Written Testimony of Mr. Al Redmer, Jr., Commissioner Maryland Insurance
- 12 -

Administration at 1 (NMHC001331). The New York Superintendent of Financial Services had
similar concerns:
DFS [Department of Financial Services] is concerned that the risk adjustment
program has created inappropriately disparate impacts among health insurance
issuers in New York and unintended consequences. Specifically, it is DFS’s
understanding that, based on the data accumulated by CMS for the upcoming
report on June 30, 2016, new and smaller issuers generally are considered to have
had relatively healthy members than their larger and more established
competitors. CMS’s anticipated determination appears to be unduly impacted by
the dates of diagnoses or recording of diagnoses of members’ medical conditions
rather than actual relative health of the members. This disparity may be because
the new and smaller health insurers have not been in operation long enough to
have amassed the long term data and records management systems that have
helped to allow the large, established health insurers to convince CMS that their
members are relatively unhealthy and, concomitantly, will allow them to receive
large payments from the risk adjustment program.
Letter from Maria T. Vullo, Superintendent of Financial Services of the State of New York, to
Sylvia M. Burwell, Secretary of Health and Human Services, and Andrew Slavitt, Acting
Administrator for the Centers for Medicare and Medicaid Services at 1-2 (dated June 28,
2016)(NMHC001335-36).
PROCEDURAL BACKGROUND
Health Connections filed its initial complaint on July 29, 2016. See Complaint for
Declaratory and Injunctive Relief, filed July 29, 2016 (Doc. 1).
subsequently filed an amended complaint.

Health Connections

See Amended Complaint for Declaratory and

Injunctive Relief at 1, filed January 12, 2017 (Doc. 21)(“Complaint”). Health Connections
alleges that HHS violated “Section 1343 of the ACA and the APA, 5 U.S.C. § 706.” Complaint
at 54. Health Connections filed the Health Connections’ Motion on April 13, 2017, see Health
Connections Motion at 1, and HHS filed the Defendants’ Motion on June 1, 2017, see
Defendants’ Motion at 1.

- 13 -

1.

The Plaintiff Mem.

Health Connections argues that using of the state average premium when calculating risk
adjustment transfer payments exceeds HHS’ authority under the ACA and is arbitrary and
capricious. See Plaintiff Mem. at 24. Health Connections observes that, under the ACA, “an
issuer may only be assessed a charge under the [risk adjustment] program ‘if the actuarial risk of
the enrollees of such plans or coverage for a year is less than the average actuarial risk of all
enrollees in all plans or coverage in such State for such year.’” Plaintiff Mem. at 24 (quoting 42
U.S.C. § 18063(a)(1)).

It follows, according to Health Connections, that “risk adjustment

assessments cannot be based on factors other than actuarial risk, and HHS is mandated to follow
this clear statutory text.” Plaintiff Mem. at 24.
Health Connections contends that HHS’ use of the state average premium when
calculating risk adjustment transfer payments is contrary to that statutory mandate, because the
state average premium “is very different than relative actuarial risk.” Plaintiff Mem. at 25.
According to Health Connections, health-insurance companies base their premiums -- and, by
extension, the state average premium -- “upon not only whether the population of insureds are
healthier or sicker, but also on whether an issuer can control its costs by paying lower prices to
hospitals and doctors, by doing a better job managing its members’ medical care, by reducing
administrative overhead, and by controlling other costs.”

Plaintiff Mem. at 25.

Health

Connections adds that using the state average premium is particularly unfair, because the state
average premium is weighted by each insurer’s marketshare, so insurers “with dominant market
positions -- such as BCBS [Blue Cross Blue Shield] in New Mexico -- drive the statewide
average premium through their own prices, which are typically quite high.” Plaintiff Mem. at
17. Health Connections contends that, instead of the state average premium, HHS should

- 14 -

calculate each insurer’s risk adjustment transfer payment using that insurer’s average premium.
See Plaintiff Mem. at 17-18.
Health Connections then addresses why, in its view, HHS’ proffered justifications for
using the state average premium are unavailing. See Plaintiff Mem. at 29. Health Connections
asserts that HHS gave two justifications for using state average premiums: (i) doing so assures
that risk adjustment is budget neutral; and (ii) doing so provides a straightforward and predicable
benchmark. See Plaintiff Mem. at 21-22. “The agency’s main point was that use of the
statewide average premium would be easy from an administrative standpoint, and the agency
could achieve budget neutrality without having to make further adjustments or calculations.”
Plaintiff Mem. at 22.
Health Connections argues that, contrary to HHS’ first justification, budget neutrality
does not justify using the state average premium, because “there is no statutory requirement that
risk adjustment be budget neutral,” i.e., that risk adjustment payments that insurers make to HHS
must equal the risk adjustment payments that HHS makes to insurers. Plaintiff Mem. at 22.
Health Connections argues that the Court should not defer to HHS’ budget-neutrality
determination, because “HHS has no specialized expertise in budgeting and appropriations, and
thus its views on budget neutrality are entitled to no deference.” Plaintiff Mem. at 22 (citing
King v. Burwell, 135 S. Ct. 2480, 2489 (2015)). Health Connections also argues that deference
is inappropriate, because “HHS has never explained why it believes the program must be budget
neutral,” and “[t]his Court owes no deference to naked assertions by agencies that lack reasoned
explanation.” Plaintiff Mem. at 22.
Health Connections continues by explaining why the Court -- when construing the
ACA’s language concerning risk adjustment for itself and not deferring to HHS -- should

- 15 -

conclude that risk adjustment need not be budget neutral. See Plaintiff Mem. at 22-23. Health
Connections notes that there is no explicit ACA language requiring budget neutrality for risk
adjustment, see Plaintiff Mem. at 22, whereas 42 U.S.C. § 18061(b)(1)(B)’s language regarding
reinsurance -- which is, essentially a temporary version of risk adjustment -- “expressly made
payments out subject to issuer’s payments in,” Plaintiff Mem. at 23.

See 42 U.S.C.

§ 18061(b)(2)(B)(“[T]he applicable reinsurance entity collects payments under subparagraph (A)
and uses amounts so collected to make reinsurance payments to health insurance issuers
described in subparagraph (A) that cover high risk individuals in the individual market . . . .”).
“The lack of such a budget neutrality provision in the risk adjustment provision of the ACA
strongly suggests that Congress intentionally omitted it and meant for the programs to be
administered differently.” Plaintiff Mem. at 23.
Health Connections also finds it significant that the risk corridors program, like the risk
adjustment program, contains no explicit budget neutrality requirement, but HHS concluded that
the risk corridor program does not need to be budget neutral, and “the GAO [Government
Accountability Office] opined that the general appropriation to HHS for carrying out its ‘other
responsibilities’ would be available for risk corridors program liabilities.” Plaintiff Mem. at 23.
Health Connections reasons that HHS could likewise fund risk adjustment through its general
appropriations if HHS implements risk adjustment in a way that is not budget neutral. See
Plaintiff Mem. at 24. Health Connections adds that, even “if HHS’s agency budget lacked
sufficient appropriations, underpaid issuers could sue in the Court of Federal Claims and recover
any unpaid monies from the Judgment Fund.” Plaintiff Mem. at 24. Health Connections also
argues, in the alternative, that even if the risk-adjustment program must be budget neutral, HHS
need not use “a formula in which it will be mathematically impossible for payments in and out to

- 16 -

ever be imbalanced,” Plaintiff Mem. at 24, because HHS could still base assessments and
payments on each issuer’s own premium, and make any necessary pro rata adjustments if there
is a shortfall of payments in,” Plaintiff Mem. at 25.
Turning to HHS’ second justification for using the state average premium when
calculating risk adjustment payments, i.e., that doing so provides a straightforward and
predicable benchmark, Health Connections asserts that HHS has “no explanation or backup data
for this statement.” Plaintiff Mem. at 25. Health Connections explains, on the contrary, that the
state average premium “is a black box for smaller issuers like NMHC,” because larger insurers’
pricing decisions “drive the statewide average.” Plaintiff Mem. at 25. Health Connections also
explains why it cannot predict its risk-adjustment liability for one year by looking at its liability
in the previous year: “NMHC must set its premiums for a given benefit year in the previous
calendar year, so that, for example, it had to finalize 2015 premiums in 2014[, but] NMHC does
not lean of its risk adjustment liability [for a given benefit year] until well into the following
year.” Plaintiff Mem. at 25-26.
Health Connections then turns its focus from the substance of HHS’ risk-adjustment
regulations to HHS’ rulemaking process. Health Connections states that, “[i]n response to
HHS’s December 2, 2015 publication of proposed rulemaking for the 2017 benefit year, NMHC
and numerous others submitted voluminous comments attacking the agency’s use of the
statewide average premium.” Plaintiff Mem. at 27-28.5 According to Health Connections, HHS
refused to respond directly to those comments when it published its final rule on March 8, 2016,
5

Health Connections explains that it, and other insurers, did not submit these comments
earlier, because “[t]he first risk adjustment results, for benefit year 2014, were published by HHS
on June 30, 2015,” and those results “made clear that the system is broken,” but by the time
those results were published “HHS had already promulgated regulations governing risk
adjustment for 2015 and 2016, maintaining the same use of the statewide average premium.”
Plaintiff Mem. at 27.
- 17 -

because that that final rule states only: “‘We did not propose changes to the transfer formula, and
therefore, are not addressing comments that are outside the scope of this rulemaking.’” Plaintiff
Mem. at 28 (quoting 2017 Final Rule, 81 Fed. Reg. at 12,230 (A.R.007774)). Again according
to Health Connections, “[t]his refusal to respond to detailed, reasoned comments from
stakeholders is the very epitome of arbitrary and capricious behavior.” Plaintiff Mem. at 28.
Health Connections observes that, when “HHS next published its proposed new rulemaking for
the 2018 benefit year on September 6, 2016,” HHS sought comments on removing some
administrative expenses from the state average premium when calculating risk-adjustment
payments. Plaintiff Mem. at 29. Health Connections also observes that HHS’ final rule for the
2018 benefit year reduces the state average premium by 14 percent when calculating risk
adjustment payments to reflect the portion of an insurer’s administrative costs that do not vary in
response to how healthy or sickly the insurer’s customers are. See Plaintiff Mem. at 30.
Health Connections maintains, however, that this modification is “too little and too late,”
because it only applies prospectively and it does not address all of Health Connections’ concerns.
Plaintiff Mem. at 30. Health Connections contends that HHS’ 2018 rule “admit[s] that it was
inflating risk adjustment assessments in 2014 and 2015 -- and will do so again for 2016 and 2016
-- by not applying this 14% adjustment,” but does nothing to address that inflation. Plaintiff
Mem. at 30. Health Connections argues that, “[a]t a minimum, if the agency has determined that
its formula was overstating actuarial risk by a calculated percentage, then that correction must be
made for all years of the program.” Plaintiff Mem. at 30. Health Connections also argues that,
notwithstanding HHS’ modification of the risk adjustment transfer formula to account for
administrative costs, the agency “ignores that there are factors driving premium levels that are
not either risk selection or administrative costs” -- such as “NMHC’s innovative medical

- 18 -

management” and its “success in securing lower prices from hospitals and doctors” -- so “HHS is
therefore still assessing charges based on factors other than actuarial risk.” Plaintiff Mem. at 31.
Health Connections next argues that, regardless of HHS’ use of state average premiums,
HHS inaccurately measures actuarial risk in the first place. See Plaintiff Mem. at 40. According
to Health Connections, HHS’ risk adjustment formula “begins by calculating a risk score for
each enrollee.” See Plaintiff Mem. at 40. Health Connections argues that the risk score reflects
the relative health, and thus, the predicated healthcare costs, of each enrollee. See Plaintiff Mem.
at 40. According to Health Connections, to calculate this risk score, an enrollee first receives a
coefficient based on age and gender. See Plaintiff Mem. at 40. It continues that the coefficient
is increased if the enrollee has been diagnosed with what HHS calls a HCC, which includes
diseases like diabetes or HIV/AIDS. See Plaintiff Mem. at 41. Essentially, according to Health
Connections, the HCC number is added to the age/gender number to calculate an enrollee’s risk
score. See Plaintiff Mem. at 41.
Health Connections believes this risk score calculation system is flawed for several
reasons. First, Health Connections argues that this system under-predicts “the costs of enrollees
who do not qualify for an HCC.” Plaintiff Mem. at 41. For example, it contends that an
individual without an HCC could still catch the flu, break a bone, or utilize preventive care
services, none of which the HCC coefficient captures. See Plaintiff Mem. at 41.

Health

Connections therefore concludes that HHS underestimates health care costs for enrollees without
an HCC. See Plaintiff Mem. at 41. According to Health Connections, this risk score calculation
produces an absurd result, because Health Connections tries to improve its members’ health so
that they do not develop HCC conditions, but if Health Connections’ enrollees do not have HCC
conditions, then Health Connections loses money. See Plaintiff Mem. at 42. Insurance carriers

- 19 -

are, therefore, according to Health Connections, given financial incentives not to improve their
enrollees’ health, which was not what Congress intended. See Plaintiff Mem. at 42.
Health Connections contends that, to fix this problem, it submitted a proposal explaining
a solution authored by former Centers for Medicare and Medicaid Services Chief Actuary Rick
Foster. See Plaintiff Mem. at 44 (citing Richard S. Foster, Method to Address Estimation Bias in
the HHS-HCC Risk Adjustment Model, at 1 (dated, July 15, 2016)(NMHC001042)(“Foster
Memorandum”)). According to Health Connections, HHS did not respond to this proposal at all.
See Plaintiff Mem. at 44 (citing 2018 Final Rule, 81 Fed. Reg. at 94,082-83 (dated December 22,
2016)(A.R.009619-20)). Health Connections contends that ignoring this proposal, submitted as
a response to HHS’ proposed rules, is arbitrary and capricious, see Plaintiff Mem. at 44 (citing
Allied Local and Regional Mfrs. Caucus v. U.S. E.P.A., 215 F.3d at 79-80), because “the
protections of notice and comment rulemaking under the APA are meaningless if the agency
were free to simply ignore comments from the public.” Plaintiff Mem. at 44-45 (citing Home
Box Office, Inc. v. F.C.C., 567 F.2d 9, 35 (D.C. Cir. 1977)).
A second flaw with HHS’ risk adjustment formula, according to Health Connections, is
that it does not accurately identify enrollees who should qualify for an HCC. Plaintiff Mem. at
45. Health Connections contends that the inaccuracy occurs for two reasons; first HHS does not
account for “partial year enrollees” and, second, HHS does not use prescription drug data when
calculating risk scores. Plaintiff Mem. at 45. Regarding the first criticism, a partial year enrollee
is one who is not enrolled in a given carrier’s health insurance plan for the full calendar year.
See Plaintiff Mem. at 45. According to Health Connections, if a partial year enrollee has an
HCC condition but receives his or her diagnosis during the part of the year in which he or she is
not on the given carrier’s plan, then “the enrollee’s risk score will be understated because the

- 20 -

plan cannot report the HCC score.” Plaintiff Mem. at 45. (citing Hickey Declaration ¶¶ 94-95,
at 22-23 (NMHC000884-85)).
Health Connections’ contends that its second criticism, HHS’ failure to use prescription
drug data, is related to the first. See Plaintiff Mem. at 46. According to Health Connections,
because individuals do not always receive an HCC diagnosis during their enrollment periods,
using prescription drug data, according to Health Connections, is a good factor to use in
calculating an enrollee’s risk score. See Plaintiff Mem. at 46. Health Connections argues that,
for example, an enrollee may have been diagnosed with diabetes before enrolling in a given
carrier’s health insurance plan. See Plaintiff Mem. at 46. According to Health Connections, that
enrollee, however, regularly fills insulin prescriptions. See Plaintiff Mem. at 46. According to
Health Connections, if HHS were to use that information, then it could accurately capture the
enrollee’s otherwise missed diagnosis. See Plaintiff Mem. at 46. Health Connections asserts
that it “is particularly hard hit by the exclusion of prescription drug data, because NMHC
prevents unnecessary hospital and physician encounters by proactively engaging with its
members to take their medications.” Plaintiff Mem. at 47.
According to Health Connections, HHS finally addressed the issues of partial year
enrollees and prescription drug data in the spring of 2016. See Plaintiff Mem. at 49. Health
Connections argues, however, that HHS did not fix the partial year enrollment problem until
2017 and will not begin to use prescription drug data until 2018. See Plaintiff Mem. at 49.
Health Connections asserts that HHS should apply these changes retroactively. See Plaintiff
Mem. at 50 (citing National Fuel Gas Supply Corp. v. F.E.R.C., 59 F.3d 1281 (D.C. Cir. 1995)).
Finally, Health Connections argues that HHS has de facto banned bronze health
insurance plans in violation of the ACA. See Plaintiff Mem. at 50. According to Health

- 21 -

Connections, in the ACA exchanges, four types of insurance plans exist. See Plaintiff Mem. at
50 (citing Hickey Declaration ¶ 73, at 17 (NMHC000879)). Health Connections contends that
bronze plans require the issuer to cover sixty percent of the insured’s health care costs, seventy
percent in silver plans, eighty percent in gold plans, and ninety percent in platinum plans. See
Plaintiff Mem. at 50 (citing Hickey Declaration ¶ 74, at 17, (NMHC000879)). A bronze plan,
therefore, has the lowest premium, but the highest deductible. See Plaintiff Mem. at 50 (citing
Hickey Declaration ¶ 74, at 17, (NMHC000879)). According to Health Connections, consumers
who do not have significant health care costs or who have limited financial resources often
purchase bronze plans, because bronze plans have the lowest premiums. See Plaintiff Mem. at
50 (citing Hickey Declaration ¶ 74, at 17 (NMHC000879)). Health Connections argues that,
“[b]ecause bronze plans are low-priced and attract a healthier population, the use of the state
average premium and the underestimation bias against healthier enrollees particularly hammer
these products.” Plaintiff Mem. at 50. Essentially, Health Connections contends that HHS’ risk
adjustment formula makes it hard for bronze plans to be profitable. See Plaintiff Mem. at 51.
Because the ACA expressly provides for bronze plans to be available, see Plaintiff Mem. at 51
(citing 42 U.S.C. 18022(d)(1)(A)), Health Connections concludes that Congress must have
intended that insurers be able to issue bronze plans without losing money. See Plaintiff Mem. at
51-52. Health Connections thus asks the Court to remand this case to the agency so that HHS
can “grapple with the question of how the agency can prevent the risk adjustment program from
gutting Congress’s intent to have viable bronze product offerings.” Plaintiff Mem. at 52. In
conclusion, Health Connections requests the Court to enter an order vacating HHS’ risk
adjustment regulations for the years 2014-2018, and order HHS to revise its regulations
consistent with the Court’s judgment. See Plaintiff Mem. at 52.

- 22 -

2.

The Defendant Mem.

HHS filed the Defendant Mem. First, HHS argues that all of Health Connections’ claims
fail, because HHS’ methodology is an “eminently reasonable and well-considered approach [to]
an exceptionally complex actuarial challenge,” and that its methodology “easily satisfies the
APA’s standard of review.” Defendant Mem. at 15. According to HHS, Health Connections
“erroneously combines its challenges to the 2014-2018 Rules in a single, multi-year attack” even
though APA review is based on “the record before the agency at the time it made its decision.”
Defendant Mem. at 18. HHS contends that the Court should conduct its analysis according to
HHS’ record at the time of the 2014 benefit year “and then proceed to consider whether the
modifications prosed in subsequent years alter that assessment for those years.” Defendant
Mem. at 18. HHS asserts that its use of the state average premium is consistent with the
statutory text and is reasonable. See Defendant Mem. at 18. It argues that § 1343 does not bar
the use of state average premiums, because that section’s only specific requirement is for the
program to assess a charge on program-eligible plans if the enrollees’ actuarial risk is less than
the state’s average actuarial risk and to make a payment to such plans if the actuarial risk of their
enrollees is greater than the state’s actuarial risk. See Defendant Mem. at 18-19. According to
HHS, Congress “did not impose any requirements as to the methodology for determining the
amounts of charges or payments.” Defendant Mem. at 19. Moreover, HHS asserts, it is not
methodologically possible to “devise a transfer formula that reflects only actuarial risk, as
NMHC suggests,” because, even if HHS could “perfectly isolate actuarial risk from other
confounding variables . . . , a formula based solely on actuarial risk would yield only a raw risk
score,” and a raw risk score “measures the expected relative cost of a particular pool of enrollees
compared to the state-wide average, but it does not predict actual expenditures.” Defendant
Mem. at 19. Consequently, HHS asserts, its methodology had to consider cost factors. See
- 23 -

Defendant Mem. at 19. HHS contends that
NMHC’s proposed alternative to the state average premium (use of a plan’s own
premium) would suffer the exact same purported flaw as the Department’s
approach: it would not “be based solely upon actuarial risk.” Rather, it would be
based on pricing choices made by individual insurance plans reflecting the very
same factors that NMHC suggests are improper, such as issuer costs,
administrative overhead, efficiency factors, and the like. But under NMHC’s
approach, risk adjustment transfers would vary based on pricing choices made by
individual plans, whereas the Department’s approach adopts a weighted average
of all such pricing in a state, thereby ensuring that the formula is uniform and
stable and minimally distorted by any extreme or inaccurate pricing decisions by
individual insurance plans. Because Congress said nothing about how risk
adjustment transfers must be calculated, NMHC’s statutory argument should be
rejected.
Defendant Mem. at 19-20 (citations omitted).
Next, HHS argues that the state average premium is not arbitrary and capricious for three
reasons. See Defendant Mem. at 20. First, plan premiums contain a risk-selection element,
because “‘healthier’ plans can charge less than ‘sicker’ plans” given that healthier members
consume less health care. Defendant Mem. at 20. HHS contends that “a risk adjustment transfer
based on a healthier plan’s lower premium might not fully capture the cost of treating sicker
enrollees or adequately compensate sicker plans for their sicker membership.” Defendant Mem.
at 20-21. Second, HHS argues that a risk adjustment charge based on a healthy plan’s lower
premium “might not adequately capture the higher cost of treating sicker members” and would,
therefore, not fulfill the program’s objective to “reduce incentives for plans to avoid high risk
enrollees.” Defendant Mem. at 21 (citing 2014 Final Rule, 78 Fed. Reg. at 15,411 (dated March
11, 2013)(A.R.0000228)). HHS argues that the state average premium more accurately measures
and distributes the costs of insuring all individuals in a risk pool. See Defendant Mem. at 21.
Third, HHS argues that, because the risk adjustment program is “self-funded and budget
neutral,” payments and charges must balance, but using a plan’s own premium for transfer
calculations would make that impossible. Defendant Mem. at 22. If healthy plans pay lower
- 24 -

charges, and sicker plans receive higher payments, HHS argues,
[b]ridging the gap between payments and charges therefore would require
one of three after-the-fact adjustments: (1) reduce payments to sicker plans, (2)
increase charges to healthier plans, or (3) split the difference between sicker and
healthier plans. NMHC does not appear to advocate or the latter two options, see
Pl.’s Mot. at 24-25, but in any event, each option has drawbacks. Reducing
payments to sicker plans would likely result in sicker plans raising their premiums
to offset the anticipated expense of their sicker membership. Increasing charges
for healthier plans would eliminate the incentives of sicker plans to control costs.
And finally, splitting the difference between healthier and sicker plans (by
increasing charges and decreasing payments) would be similar to using the state
average premium, but it would require an after-the-fact adjustment that would not
be known until the program year concluded.
Defendant Mem. at 22-23 (citations omitted). HHS contends that, given the advantages
of using state average premiums, “[t]he record thus amply demonstrate[s] that the Department
considered the relevant policy choices and rationally elected to use a state-wide average.”
Defendant Mem. at 23.
HHS disputes Health Connections’ arguments in opposition to using state average
premiums. See Defendant Mem. at 23-26. First, HHS contends that using the state average
premium does not encourage gaming by large insurers, because HHS’ transfer formula “does not
directly reflect a plan’s actual premiums at all; rather, it calculates the difference between the
plan’s expected costs with risk selection and the plan’s expected costs without risk selection,
using the state average premium on both sides of the equation as an estimation of average cost.”
Defendant Mem. at 23-24 (emphasis in original). HHS contends that its approach “neither
penalizes cost-cutting nor rewards inefficiency,” but rather “strikes a middle ground” by
assuming an average level of efficiency. Defendant Mem. at 24. According to HHS, Health
Connections’ proposed system “would encourage sicker plans to charge higher premiums to
increase their payments and healthier plans to charge lower premiums to reduce their charges”
because “plans with the same risk score would owe or receive different amounts based on

- 25 -

individual pricing decisions.” Defendant Mem. at 24. Second, HHS asserts that using the state
average premium does not penalize cost-cutting plans, because its regulations require providing
advance notice of risk adjustment formula so that an issuer can “price any expected payments or
charges into their rates.”

Defendant Mem. at 24.

Third, HHS contends that it provided

considerable information, data, and research related to its decision, which “amply demonstrate[s]
the Department’s rationale for adopting the state average premium.” Defendant Mem. at 25.
Additionally, HHS contends that it must “only provide a ‘rational connection between the facts
found and the choice made,’ such that the ‘path may reasonably be discerned.’” Defendant
Mem. at 26 (quoting Encino Motorcars, LLC v. Navarro, 136 S. Ct. at 2125.)
Next, HHS contends that its use of HCCs is reasonable. See Defendant Mem. at 26-28.
HHS asserts that its risk adjustment model is not meant to transfer risk of random events like
accidents, but rather is meant to compensate plans for enrollees’ predicable medical conditions
that could influence enrollment decisions. See Defendant Mem. at 27. HHS also contends that
its model already incorporates costs of treating random events. See Defendant Mem. at 27-28.
HHS also argues that its approach to capturing HCCs is reasonable, because, HHS contends,
HHS “considered and reasonably addressed” whether to adjust its methodology for partial year
enrollment and incorporate prescription drug data. Defendant Mem. at 28-29. See Defendant
Mem. at 29-34. HHS further contends that the program’s approach to its bronze plans is
reasonable. See Defendant Mem. at 34-37. HHS asserts that it “grappled with” the relationship
between state exchange actuarial values and its risk adjustment program, ultimately “adopting
different risk score models for each metal level plan and catastrophic plan.” Defendant Mem. at
35. HHS explains that it
also included an adjustment for actuarial value in the transfer formula so that the
program does not compensate plans for differences in actuarial value that are

- 26 -

already reflected in the premiums charged by such plans. However, the
Department reasonably elected not to adopt separate risk pools for the
different metal level plans because “this approach would fail to correct for
systematic risk selection across ‘metal levels[.’] That is, low risk enrollees
would tend to migrate to plans with a lower actuarial value . . . which would then
gain a premium advantage attributable to risk selection. This result would not
address the mandate of the ACA, which requires that transfer payments be made
between plans based on the[] actuarial risk of their enrollees.” Thus, to the
extent NMHC suggests that the Department has not already exhaustively
“grappled” with the relationship between metal levels and risk adjustment, it is
wrong.
Defendant Mem. at 35 (citations omitted)(quoting State Health Insurance Exchange Risk
Adjustment and Plan Metals Level at 6 (dated December 15, 2011)(A.R.000814-15). HHS also
disputes Health Connections’ contention that HHS’ methodology is arbitrary and capricious for
not relieving bronze plans of the financial consequences of risk adjustment, arguing that:
(i) bronze plans typically have healthier enrollees, and § 1343 requires those enrollees to pay risk
adjustment charges; and (ii) “administrative review is not based on hindsight and it does not
appear that Health Connections raised this outcome-oriented critique until the 2018 rulemaking.”
Defendant Mem. at 36.
HHS also contends that the 2015-2017 rules are consistent with the statute and are
reasonable. See Defendant Mem. at 36. According to HHS, it was not arbitrary and capricious
to not respond to comments addressing issues beyond the scope of proposed rules, because HHS
was “not obligated to reconsider methodological choices it already had exhaustively considered
or to respond anew to comments questioning those choices.” Defendant Mem. at 36.
Next, HHS asserts that the 2018 rule is consistent with the statute and is reasonable. See
Defendant Mem. at 37-40.

HHS notes that it has addressed many of Health Connections

complaints by adopting a downward adjustment to the state average premium and preventative
health costs, including additional partial year enrollment factors, and making limited use of

- 27 -

prescription drug data. See Defendant Mem. at 37. As to Health Connections’ “remaining
grievances,” HHS asserts that: (i) the adjustment to the state average premium is reasonable,
because “there is nothing arbitrary and capricious about using a mean to approximate overall
health costs in a state nor does section 1343 require risk adjustment transfers” to certain
individuals, Defendant Mem. at 38; (ii) HHS reasonably addressed a particular proposed formula
adjustment concerning estimation bias; and (iii) its methodology does not make it impossible for
bronze plans to be profitable, and its internal process to improve its models means that “judicial
relief is unnecessary,” Defendant Mem. at 40-41.
HHS also argues that no basis for Health Connections’ requested relief exists. See
Defendant Mem. at 41-44. First, HHS asserts that the Court lacks jurisdiction to award primarily
monetary relief. See Defendant Mem. at 41. HHS contends that “although NMHC nominally
seeks an injunction requiring the Department to revise its risk adjustment formula . . . the thrust
of its suit is for a refund of money already paid to the Department.” Defendant Mem. at 41.
Second, HHS argues that, even if the sought refunds are not considered money damages,
“vacatur should still be denied because vacating the risk adjustment methodology for all prior
years would harm plans that enrolled sicker than average enrollees.” Defendant Mem. at 43.
3.

The Plaintiff Reply.

Health Connections replied. See Plaintiff’s Reply and Opposition to Defendant’s CrossMotion for Summary Judgment at 1, filed July 13, 2017 (Doc. 40)(“Plaintiff’s Reply”). Health
Connections first argues that HHS “has no discretion to ignore real world developments.”
Plaintiff’s Reply at 8. Specifically, Health Connections contends that “under HHS’s logic
because the only rulemaking that should be reviewed is the original one for the 2014 benefit year
. . . the only evidence the Court should review is what was before the agency in 2012/2013

- 28 -

during that first rulemaking.” Plaintiff’s Reply at 8 (citing Defendant Mem. at 12-13, 17-18, 27,
36-37). Health Connections contends that this logic is incorrect, because “‘when there is a
known or significant change in the data underlying an agency decision, the agency must either
take that change or trend into account, or explain why it relied solely on data pre-dating that
change or trend.’” Plaintiff’s Reply at 8 (quoting Zen Magnets, LLC v. Consumer Prod. Safety
Comm’n, 841 F.3d 1141, 1149 (10th Cir. 2016)). Health Connections asserts that HHS knew
about the flaws in its risk adjustment formula shortly after the program began, but did not change
the formula. See Plaintiff’s Reply at 9-10. Health Connections thus avers that, when an agency
ignores new data without adequate explanation, which it believes HHS did, such conduct is
arbitrary and capricious. See Plaintiff’s Reply at 8-9 (citing Magnets, LLC v. Consumer Prod.
Safety Comm’n, 841 F.3d at 1149).
Health Connections next re-asserts its argument that HHS’ use of the state average
premium in its risk adjustment formula violates the ACA. See Plaintiff’s Reply at 12. Health
Connections argues that, because the ACA provides that the risk adjustment program should
assess a charge on insurers based on actuarial risk, see Plaintiff’s Reply at 12 (citing 42 U.S.C. §
18063(a)(1)-(2)), and HHS assesses a charge based on “multiplying relative actuarial risk against
the statewide weighted average premium,” HHS violates the ACA. See Plaintiff’s Reply at 12.
Health Connections adds that nothing in the ACA’s text indicates charges and payments should
be based on any factor other than actuarial risk. See Plaintiff’s Reply at 13 (citing 42 U.S.C. §
18063(a)(1)-(2)).
Health Connections then re-asserts that HHS’ use of state average premiums is arbitrary
and capricious. See Plaintiff’s Reply at 15. Specifically, HHS contends that the one factor
which Congress directed HHS to consider in developing a risk adjustment program was actuarial

- 29 -

risk, but HHS’ use of state average premiums considers other factors unrelated to actuarial risk,
such as “how effectively a plan negotiates prices with hospitals and physicians, how well it
manages its members’ medical care, and how ably it controls administrative expenses.”
Plaintiff’s Reply at 16. Health Connections adds that, contrary to HHS’ assertion, premiums are
not a proxy for actuarial risk. See Plaintiff’s Reply at 18 (citing Defendant Mem. at 16-21).
Health Connections then responds to HHS’s contention that the use of state average
premiums “‘reduce[s] incentives for plans to avoid high risk enrollees.’” Plaintiff’s Reply at 19
(quoting Defendant Mem. at 21). To this contention, Health Connections rejoins that the ACA
has separate statutory provisions “forbidding carriers from denying coverage or raising
premiums for sicker enrollees,” Plaintiff’s Reply at 19 (citing 42 U.S.C. § 300gg-3), and that
“nothing in the text of the ACA . . . suggests that Congress intended risk adjustment to somehow
be the enforcement mechanism for these provisions.”

Plaintiff’s Reply at 19.

Health

Connections then largely repeats its arguments that HHS’ justifications for state average
premiums, i.e., budget neutrality and predictability, are flawed. See Plaintiff’s Reply at 20-21.
Next, Health Connections avers that “HHS fails to account for the actual health care costs
of healthier enrollees.” Plaintiff’s Reply at 21. Health Connections asserts that HHS “nowhere
addresses the evidence that the formula just flat out does not work in predicting the costs of
medical care.” Plaintiff’s Reply at 22. Health Connections adds that HHS failed to respond to a
comment from Health Connections about this problem and that, when an agency receives
“‘critical commentary,’” it must “‘respond in a reasoned manner’” to that comment. Plaintiff
Reply. at 23-24 (quoting FMBE Bank Ltd. v. Lew, 209 F. Supp. 3d 299, 333 (D.D.C. 2016)).
Health Connections then re-alleges that HHS “violates Congressional Intent to have a
robust market in bronze plans.”

Plaintiff’s Reply at 24.

- 30 -

Health Connections repeats that

Congress intended for bronze plans to be available in the ACA exchanges, but that HHS’ risk
adjustment formula makes it difficult for bronze plans to be profitable. See Plaintiff’s Reply at
24. Health Connections adds that HHS did not respond to Health Connections’ comments on
this point during the 2018 rulemaking period. See Plaintiff’s Reply at 24.
Health Connections then re-asserts its arguments that HHS wrongfully excluded
prescription drug data from its risk adjustment formula before 2018, see Plaintiff’s Reply at
25-27, and that HHS’ formula does not account for partial year enrollees, see Plaintiff’s Reply at
27-28, before addressing a new point -- HHS’ jurisdictional argument, see Plaintiff’s Reply at
28. Health Connections asserts that “HHS contends that NMHC seeks only money damages for
past risk adjustment calculations and thus this action belongs in the Court of Federal Claims.”
Plaintiff’s Reply at 28. Health Connections rejoins that this argument is meritless, because “the
Prayer for Relief in NMHC’s Amended Complaint requests only declaratory and injunctive relief
. . . [and] asserts only one count under the APA, which only permits declaratory and injunctive
relief.” Plaintiff’s Reply at 28 (alteration added)(citing Complaint at 56; 5 U.S.C. §§ 702, 706).
Health Connections also emphasizes that it seeks only equitable relief, and not damages, for its
claims relating to the years 2014-2016.

See Plaintiff’s Reply at 24.

Specifically, Health

Connections asserts that it does not ask the Court to refund any charges for those years, but
rather, to invalidate regulations used during those years so that HHS can then fix its regulatory
scheme. See Plaintiff’s Reply at 29. Health Connections then avers that, if under HHS’ fixed
scheme, HHS owes back money to Health Connections, either HHS can pay Health Connections
or Health Connections can sue in the Court of Federal Claims. See Plaintiff’s Reply at 29.
Health Connections concludes that the Court should grant its motion for summary judgment,
deny HHS’ cross-motion, and vacate HHS’ risk adjustment regulations. See Plaintiff’s Reply at

- 31 -

30. Alternatively, Health Connections requests that the Court remand the matter to HHS without
a vacatur.6
4.

The Defendants’ Reply.

HHS replied to the Plaintiff’s Reply. See Defendants’ Reply in Support of Cross-Motion
for Summary Judgment, filed August 17, 2017 (Doc. 41)(“Defendants’ Reply”). First, HHS
argues that, to prevail, Health Connections must show that Congress unambiguously and directly
addressed how HHS should measure costs in its transfer formula, but that Congress has offered
no such directive. See Defendants’ Reply at 3-4. HHS contends that, contrary to Health
Connections’ assertions: (i) HHS’ transfer formula incorporates actuarial risk; (ii) the important
question is whether Congress unambiguously and directly prohibits HHS from using weighted
average premiums to approximate costs; (iii) it is irrelevant whether the state average premium
inflates risk adjustment charges; and (iv) Health Connections paying a large percentage of its
premiums in risk adjustment charges is consistent with Congressional intent. See Defendants’
Reply at 4-5. HHS revisits its argument that HHS’ use of a state average premium is reasonable,
because it “preserves incentives to control costs while also meeting the health needs of sicker
enrollees.” Defendants’ Reply at 6. See Defendants’ Reply at 6-10.
Second, HHS argues that its use of HCCs is reasonable and that HHS adequately
responded to concerns about estimation bias. See Defendants’ Reply at 11-15. HHS argues that
6

Health Connections later filed a short Notice of Supplemental Authority, filed
September 18, 2017 (Doc. 44)(“Notice”). The notice mentioned a recent case, American College
of Emergency Physicians v. Price, 264 F. Supp. 3d 89 (D.D.C. 2017)(Kollar-Kotelly, J.), in
which the United States District Court for the District of Columbia held that HHS acted
arbitrarily and capriciously “by failing to ‘seriously respond’ to public comments regarding a
proposed rule pertaining to” the ACA. Notice at 1 (quoting American College of Emergency
Physicians v. Price, 264 F. Supp. 3d at 94). Health Connections briefly asserts that American
College of Emergency Physicians v. Price is analogous to this case. See Notice at 1 (“Just like
Price, this case involves, inter alia, APA claims challenging HHS’ failure to meaningfully
respond to public comments regarding rules issued pursuant to the ACA.”).
- 32 -

(i) Health Connections impermissibly bases its arguments on information that did not exist
before HHS enacted the rule that Health Connections challenges, see Defendants’ Reply at
11-12; (ii) Health Connections’ challenge to the 2017 and 2018 plan year models is flawed,
because HHS added preventative services to its models to improve predictability vis-à-vis nonHCC enrollees, see Defendants’ Reply at 12; and (iii) HHS adequately addressed the former
Chief Actuary of the Centers for Medicare and Medicaid Services, Richard Foster’s, proposal to
adjust plan liability risk scores, see Defendants’ Reply at 13-15.
Third, HHS again argues that its approach to identifying risk has been reasonable in all
relevant years. See Defendants’ Reply at 15-19. HHS contends that: (i) it thoroughly considered
whether to incorporate pharmacy data in its model as a measure of HCC status, seeking
comments conducting analyses, see Defendants’ Reply at 15-18; and (ii) it reasonably addressed
partial year enrollment and Health Connections “has identified nothing that would have led the
Department to believe that [its approach was] insufficient, much less unreasonable, to address
partial year enrollment,” Defendants’ Reply at 18-19.
Fourth, HHS asserts that its approach to bronze plans is reasonable. See Defendants’
Reply at 19-20. HHS contends that: (i) Health Connections has not provided evidence that “risk
adjustment is driving bronze plans out of existence,” Defendants’ Reply at 19; (ii) Health
Connections “fails to challenge any specific action or inaction” of HHS, Defendants’ Reply at
19; and (iii) HHS is “already monitoring the effects of the program on bronze plans, and,
therefore, NMHC’s requests for the Department to ‘grapple’ with the issue is moot,” Defendants’
Reply at 20.
Fifth, HHS asserts that Health Connections’ remaining arguments fail. See Defendants’
Reply at 20-24. HHS argues that: (i) it did not ignore criticisms against the program, but rather

- 33 -

analyzed results and sought public comments to improve it, see Defendants’ Reply at 20; (ii) it
properly “declin[ed] to retroactively modify rules that had already been promulgated,”
Defendants’ Reply at 20-21; and (iii) Health Connections’ suggestions that HHS’ risk adjustment
program created broader problems in the ACA market are irrelevant and meritless, see
Defendants’ Reply at 23-24.
Sixth, and finally, HHS argues that the Court should deny retroactive relief.

See

Defendants’ Reply at 24-25. HHS contends that it has “already adopted the majority of the
reforms that NMHC has requested.” Defendants’ Reply at 24-25. According to HHS, Health
Connections seeks in part “to vacate the program back to its inception,” which means that it is
“reasonable to infer that NMHC’s primary motivation for bringing this case is not to correct the
program moving forward, but to obtain money for past years.” Defendants’ Reply at 24. HHS
also argues that, even if Health Connections’ claim is proper under the APA, “NMHC has not
met its burden to show that the relief it seeks is either equitable or achievable.” Defendants’
Reply at 25.7
5.

The Hearing.
The Court held a hearing. See Draft Transcript of Motion Proceedings (taken

January 22, 2018)(“Tr.”).8 Health Connections took to the podium first and argued that HHS’

7

HHS also responds to Health Connections’ Notice of Supplemental Authority. See
Defendants’ Response to Plaintiff’s Notice of Supplemental Authority, filed September 22, 2017
(Doc. 45). HHS argue that American College of Emergency Physicians v. Price does not help
Health Connections’ position, because that case determined that the agencies did not seriously
respond to concerns about its methodology and ignored proposed alternative methodologies,
whereas HHS seriously considered all the issues raised in this case and has not ignored any
alternative proposals. See Defendants’ Response to Plaintiff’s Notice of Supplemental Authority
at 1-2 (citing American Coll. of Emergency Physicians v. Price, 2017 WL 3836045, at *4).
8

The Court’s citations to the hearing transcript refer to the court reporter’s original,
unedited version. Any final transcript may contain slightly different page and/or line numbers.
- 34 -

use of a statewide premium average in its actuarial risk calculation does not accurately capture
actuarial risk, so HHS’ use of that premium average violates 42 U.S.C. § 18063’s plain language.
See Tr. at 4:17-23 (Bassman, Court); id. at 5:6-9 (Bassman). Health Connections continued that
the statute is not lengthy, but its command is “pretty clear”: an issuer makes a payment if the
“actuarial risk is below average in this state.” Tr. at 11:18-22 (Bassman). Health Connections
explained that there are two steps to HHS’ actuarial risk calculation: (i) HHS calculates your
relative actuarial risk by measuring whether your enrollees’ health is above or below the state
average; and (ii) HHS multiplies that amount by the statewide average premium. See Tr. at 5:1422.

Health Connections contends that both steps of that two-step process are arbitrary,

capricious, and contrary to law. See Tr. at 5:22-25 (Bassman).
The Court challenged Health Connections’ characterization of the statute and
noted that the statute allows HHS to consider factors beyond actuarial risk. See Tr. at 13:15-20
(Court). Health Connections rejoined that “the statutory text . . . speaks of actuarial risk” and not
of other factors. Tr. 14:11-13 (Bassman). It also argued that the proper premium against which
to measure is the issuer’s own premium, because an issuer’s own premium captures efficiencies
and costs more effectively. Tr. at 15:1-4 (Bassman).
The Court asked whether using the issuer’s premiums “undercut[s] your statutory
language argument,” because it signals that HHS’ use of the statewide average premium
calculation is not arbitrary and capricious. See Tr. at 15:6-18 (Court). Health Connections
rejoined that Congress’ “mandate is to get as close to actuarial risk as you can” and that a
healthcare issuer’s own premium is a far closer approximation of actuarial risk than the statewide
average premium. Tr. at 15:22-16:4 (Bassman). Health Connections added, however, that, even
setting the statute aside, HHS was not reasonable in using a statewide average. See 16:9-12

- 35 -

(Bassman)(“[T]hat’s the second part of our attack . . . .”). The Court rejoined that the analysis
for which Health Connections is looking for is “much more rigorous . . . than [] courts normally
do” under the APA. Tr. at 16:17-22 (Court). Health Connections argued that, under arbitrary
and capricious review, HHS must look at “appropriate factors in making its rational decisions,”
and Health Connections contended that HHS did not analyze actuarial risk as a factor, the one
factor it had to consider under the statute. Tr. at 17:2-14 (Bassman).
Responding to HHS’ argument from the briefing that using an issuer’s own premium
incentivizes manipulating the premium, Health Connections argued that it could not manipulate
its rates for beneficial effect, because: (i) issuers cannot unilaterally set their prices; and (ii) the
federal and state governments may disapprove premium if they are too high or too low. See Tr.
at 17:21-18:11 (Bassman).

Health Connections added that manipulation would not work,

because the risk adjustment charge is unpredictable, based on their enrollees’ health and their
competitor’s premiums. See Tr. at 18:21-19:11 (Bassman).
The Court asked whether the predictability issue was a “flaw in the statute.” Tr. at
20:20-21 (Court). Health Connections admitted that “certainly [the] predict[ability] problem is
partly baked into the statute.” Tr. at 20:23-24 (Bassman). The Court responded that if the
predictability problem is in the statute, how could HHS have been acting arbitrarily and
capriciously. Tr. at 21:2-3 (Court). Health Connections rejoined, somewhat circularly, that HHS
could still have acted arbitrarily and capriciously, because there is a better method of calculating
actuarial risk: the “issuers own premium.” Tr. at 21:4-7 (Bassman).
Health Connections continued that the second reason HHS has acted arbitrarily and
capriciously is that, in promulgating their rules, HHS reasons that it should act in a budget
neutral way. See Tr. at 21:17-19 (Bassman). Health Connections argued that budget neutrality

- 36 -

is not an appropriate factor upon which to rely, because the statute is not budget neutral. Tr. at
22:4 (Bassman). Health Connections argued that several sections in the statute are expressly
budget neutral, but the risk adjustment section is not, so Congress could not have intended for the
risk adjustment provision to be budget neutral. See Tr. at 22:12-23:5 (Bassman).

The Court

responded that “budget neutrality is at least reasonable. It may not be the best reading of the
statute, but [it] seems . . . at least reasonable.” Tr. at 25:4-9. Health Connections rejoined that
budget-neutrality could not be reasonable, because Congress appropriated bridge funds, and, in
other programs, such as reinsurance, HHS accounted for reductions in some years which extra
funds would make up for in later years.

See Tr. at 26:23-27:4 (Bassman); id. at 28:1-6

(Bassman).
HHS responded. See Tr. at 35:14-15 (Powers). It argued that the statute gives “broad
discretion to determine the methods and standards . . . applicable to the risk adjustment
activities.” Tr. at 39:24-40:1 (Powers). It added that it used the statewide average premium,
because “it reflects the cost of insuring [the] sickest people” as it reflects all premiums in the
calculation.

Tr. at 40:9-13 (Powers).

It continued that “[s]tatewide average premium

incorporates elements of actuarial risk selection, including administrative expense contained
there, and so it is consistent with the status to use statewide average premium[s] as a cost factor
in conjunction with the risk score.” Tr. at 40:19-24 (Powers).
HHS admitted that the “statute does not require budget neutrality,” Tr. at 44:3-5
(Powers), but it contended, however, that requiring budget neutrality is a reasonable
interpretation of the statute given that the agency is “not permitted to obligate the Government in
the absence of appropriations,” Tr. at 44:8-16 (Powers). HHS also argued that it was likely that
Congress thought this should be budget neutral, because “Congress in the first instance assumed

- 37 -

that states would appropriate these risk adjustment program[s],” and it would be “quite strange to
think that Congress was obligating [states] . . . to make up the shortfall or difference in payments
and transfers.” Tr. at 45:14-20 (Powers).
HHS noted Health Connections’ argument that taking into account all premiums “sweeps
in other factors more than actuarial risk,” but contended that “us[ing] an issuers own premium”
sweeps in the same non-actuarial risk factors” Tr. at 46:23-47:6 (Powers). HHS contended that
Health Connections calculation is not focused on actuarial risk, as the statute commands, but is
focused on “efficiency.” Tr. at 47:9 (Powers). HHS argued that “competition and efficiency” is
not present in the risk adjustment portion of the statute. Tr. at 48:8-10 (Powers)(citing 42 U.S.C.
§ 18063). It acknowledged that the “statute has a goal of increasing access to quality and
affordable health insurance, so that contains elements of both . . . increasing competition and
therefore bringing down prices, but also assuring that, regardless . . . of health status, [people]
have [access] to health plans that can satisfy their health needs.” Tr. at 48:12-18 (Powers). HHS
further conceded that there is “nuance” in the statute, but that “[t]he agency is entitled in the first
instance to w[eigh] these various policies particularly where the statute just gives the agency
broad discretion to spe[cify] standards applicable for the program.” Tr. at 48:22-49:2 (Powers).
Health Connections took the podium again and argued that HHS’ methodology for
calculating the enrollees’ risk scores is arbitrary and capricious. See Tr. at 61:4-9 (Bassman). It
argued that each enrollee’s risk score is calculated with three inputs: age, gender, and an HCC
code, which is given to an enrollee if he or she has a serious chronic condition. See Tr. at
61:9-15 (Bassman). Health Connections argued that the HCC code is defective, because it
misses a lot of people who have serious conditions, but do not qualify for an HCC code, such as
people with chronic lower back pain. See Tr. at 61:16-22 (Bassman). Health Connections

- 38 -

continued that people who have chronic diseases who do not have an HCC code cost an issuer
just as much as those with an HCC code. See Tr. at 62:2-13 (Bassman). From that analysis,
Health Connections argues that HHS’ calculation is miscalculating the actuarial risk, because it
does not account for all enrollees and “underestimates the cost of enroll[ees] without an HCC by
31 percent.” Tr. at 62:19-63:11 (Bassman).
HHS rejoined that the risk adjustment program is meant to “deal with systematic risks
that [are] predictable.” Tr. at 72:9-22 (Powers). HHS also noted that it had received many
comments “supportive of the various . . . logical choices” it had made. Tr. at 73:16-18 (Powers).
It also argued that “statistical analyses that the agency has applied to its methodology indicate
that it is performing just as well if not better than similar kinds of commercially available risk
adjustment methodologies.” Tr. at 73:21-25 (Powers).
Health Connections next contended that there is no risk of the issuers “gaming” the
system. Tr. at 82:6-8 (Bassman). It argued that there is no evidence in the administrative record
demonstrating that issuers pressured doctors into prescribing medications not needed so the
issuer could obtain an HCC code and a concomitant financial benefit. See Tr. at 82:9-83:6
(Bassman). It also argued that doctors are pressured from “making unnecessary prescriptions,”
as they could face sanctions from their state. Tr. at 83:9-15 (Bassman).
The Court subsequently asked HHS what the effect would be if it determined HHS’
exclusion of data was arbitrary and capricious. See Tr. at 88:14-17 (Court). HHS responded that
several issuers would ask the agency for a refund, but that HHS no longer had that money,
because “it’s been paid out to the recipients under the program.” Tr. at 88:20-25 (Powers). HHS
argued that this scenario would cause chaos for many issuers, because issuers might think that
HHS would “try to take the money back from them.” Tr. at 89:1-6 (Powers). See Tr. at 89:9-12

- 39 -

(Powers)(“[I]f your honor was to go back retroactively and vacate prior rules . . . that would have
some significant and . . . bad effects on this market.”).

Responding to this colloquy, Health

Connections noted that it was within the Court’s power to invalidate improper rules promulgated
in the past, see Tr. at 97:24-98:2 (Court), and added that, if the Court were to rule on the 2017
risk adjustment rule, such a ruling would be “entirely prospective,” Tr. at 97:6 (Bassman).
Health Connections then argued that HHS’ rules are biased against the bronze plans. See
Tr. at 105:20-23 (Bassman).

It argued that bronze plans attract healthier people, but,

consequently, issuers that offer the bronze plan are always hit with a risk adjustment charge. See
Tr. at 106:6-107:7 (Bassman). Health Connections concludes that, if HHS does not change the
statewide premium average calculation, something has to be done to “allow bronze plans to be
viable and functioning.” Tr. at 107:16-19 (Bassman).
HHS rejoined that the APA requires a more specific attack and that “the bronze plan
point is accounted for already to some extent in the formula.” Tr. at 109:8-12 (Powers). HHS
also argued that “it’s not clear what effective rel[ief] the Court could actually offer” given that
HHS is already making changes to try and address “certain subpopulations” including those
using the bronze plan. Tr. at 109:17-25 (Powers). HHS also contended that it is not clear yet
that the risk adjustment formula “renders bronze plans . . . uneconomic.”

Tr. at 110:1-2

(Powers). It added that there had not been a “widespread problem with . . . bronze plans” as “a
result of” agency action, so the risk adjustment program has not had a clear adverse effect on the
bronze plans. Tr. at 110:16-22 (Powers).
Health Connections responded that the bronze plan issue would be cured if the Court
addressed either the statewide average premium or the estimation bias issue. See Tr. at 112:1-8
(Bassman). It added that, if the Court did not address either of those issues, it would ask the

- 40 -

Court to order HHS to respond to Health Connections’ comment about how HHS was going to
adjust the formula so bronze plans are not unfairly penalized. See Tr. at 112:9-15 (Bassman).
Finally, Health Connections argued that the Court has jurisdiction, even though HHS
argues that the Health Connections’ claims are backward-looking so are subject solely to the
Court of Federal Claims’ jurisdiction, because: (i) they ask for prospective relief on HHS’
2017-18 rule; and (ii) the relief requested for the 2014-16 rules is not monetary. See Tr. at
116:22-117:16 (Powers). Health Connections admits that, if a new rulemaking results from the
Court’s ruling, Health Connections “may be owed money under it.” Tr. at 117:20-22 (Bassman).
It added, however, that it is not clear under established case law if a refund is money damages.
See Tr. at 118:1-2 (Bassman).
In response, HHS noted that, as to the 2017 and 2018 rules, it is not contesting
jurisdiction. See Tr. at 119:1-2 (Powers). HHS argued, however, that Health Connections’ case
“is fundamentally seeking money damages from the Government” and “the APA does not waive
sovereign immunity to do that.” Tr. at 120:5-8 (Powers). HHS concluded that, if the Court
determines that there is jurisdiction and that Health Connections’ claims have merit, the Court
should remand the matter to the agency without vacating any rules. See Tr. at 123:1-6 (Powers).
LAW REGARDING RULE 12(b)(1)
“Federal courts are courts of limited jurisdiction; they are empowered to hear only those
cases authorized and defined in the Constitution which have been entrusted to them under a
jurisdictional grant by Congress.” Henry v. Office of Thrift Supervision, 43 F.3d 507, 511 (10th
Cir. 1994)(citations omitted). Plaintiffs generally bear the burden of demonstrating a court’s
jurisdiction to hear his or her claims. See Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83,
104 (1998)(“[T]he party invoking federal jurisdiction bears the burden of establishing its

- 41 -

existence.”). Rule 12(b)(1) of the Federal Rules of Civil Procedure allows a party to challenge
subject-matter jurisdiction by motion. See Fed. R. Civ. P. 12(b)(1). The United States Court of
Motions to dismiss for lack of subject-matter jurisdiction “generally take one of two forms: (1) a
facial attack on the sufficiency of the complaint’s allegations as to subject-matter jurisdiction; or
(2) a challenge to the actual facts upon which subject matter jurisdiction is based.” Ruiz v.
McDonnell, 299 F.3d 1173, 1180 (10th Cir. 2002).
On a facial attack, a plaintiff is afforded safeguards similar to those provided in
opposing a rule 12(b)(6) motion: the court must consider the complaint’s
allegations to be true. See Ruiz v. McDonnell, 299 F.3d at 1180; Williamson v.
Tucker, 645 F.2d 404, 412 (5th Cir. 1981). But when the attack is aimed at the
jurisdictional facts themselves, a district court may not presume the truthfulness
of those allegations. A court has wide discretion to allow affidavits, other
documents, and a limited evidentiary hearing to resolve disputed jurisdictional
facts under Rule 12(b)(1). In such instances, a court’s reference to evidence
outside the pleadings does not convert the motion to a Rule 56 motion.
Hill v. Vanderbilt Capital Advisors, LLC, F. Supp. 2d 1228, 1240-41 (D.N.M. 2011)(Browning,
J.)(quoting Alto Eldorado Partners v. City of Santa Fe, 2009 WL 1312856, at *8-9 (D.N.M.
March 11, 2009)(Browning, J.)). See New Mexicans for Bill Richardson v. Gonzales, 64 F.3d
1495, 1499 (10th Cir. 1995); Holt v. United States, 46 F.3d 1000, 1003 (10th Cir. 1995). Where,
however, the court determines that jurisdictional issues raised in a rule 12(b)(1) motion are
intertwined with the case’s merits, the court should resolve the motion under either rule 12(b)(6)
of the Federal Rules of Civil Procedure or rule 56 of the Federal Rules of Civil Procedure. See
Franklin Sav. Corp. v. United States, 180 F.3d 1124, 1129 (10th Cir. 1999); Tippett v. United
States, 108 F.3d 1194, 1196 (10th Cir. 1997). “When deciding whether jurisdiction is intertwined
with the merits of a particular dispute, ‘the underlying issue is whether resolution of the
jurisdictional question requires resolution of an aspect of the substantive claim.’” Davis ex rel.
Davis v. United States, 343 F.3d 1282, 1296 (10th Cir. 2003)(quoting Sizova v. Nat’l Inst. of
Standards & Tech., 282 F.3d 1320, 1324 (10th Cir. 2002)).
- 42 -

LAW REGARDING JUDICIAL REVIEW OF AGENCY ACTION
Under the APA,
[a] person suffering legal wrong because of agency action, or adversely affected
or aggrieved by agency action within the meaning of a relevant statute, is entitled
to judicial review thereof. An action in a court of the United States seeking relief
other than money damages and stating a claim that an agency or an officer or
employee thereof acted or failed to act in an official capacity or under color of
legal authority shall not be dismissed nor relief therein be denied on the ground
that it is against the United States or that the United States is an indispensable
party. The United States may be named as a defendant in any such action, and a
judgment or decree may be entered against the United States: Provided, that any
mandatory or injunctive decree shall specify the Federal officer or officers (by
name or by title), and their successors in office, personally responsible for
compliance. Nothing herein (1) affects other limitations on judicial review or the
power or duty of the court to dismiss any action or deny relief on any other
appropriate legal or equitable ground; or (2) confers authority to grant relief if any
other statute that grants consent to suit expressly or impliedly forbids the relief
which is sought.
5 U.S.C. § 702. The APA states that district courts can:
(1)

compel agency action unlawfully withheld or unreasonably delayed; and

(2)
hold unlawful and set aside agency action, findings, and conclusions found
to be-(A)
arbitrary, capricious, an abuse of discretion, or otherwise
not in accordance with law;
(B)
contrary to constitutional right, power, privilege, or
immunity;
(C)
in excess of statutory jurisdiction, authority, or limitations,
or short of statutory right;
(D)

without observance of procedure required by law;

(E)
unsupported by substantial evidence in a case subject to
sections 556 and 557 of this title or otherwise reviewed on the
record of an agency hearing provided by statute; or
(F)
unwarranted by the facts to the extent that the facts are
subject to trial de novo by the reviewing court.
5 U.S.C. § 706.

- 43 -

Under Olenhouse, 42 F.3d at 1560, “[r]eviews of agency action in the district courts
[under the APA] must be processed as appeals. In such circumstances the district court should
govern itself by referring to the Federal Rules of Appellate Procedure.” 42 F.3d at 1580. See
Wildearth Guardians v. U.S. Forest Serv., 668 F. Supp. at 1323. “As a group, the devices
appellate courts normally use are generally more consistent with the APA’s judicial review
scheme than the devices that trial courts generally use, which presume nothing about the case’s
merits and divide burdens of proof and production almost equally between the plaintiff and
defendant.” Northern New Mexicans Protecting Land and Water Rights v. United States, 2015
WL 8329509, at *9 (D.N.M. 2015)(Browning, J.).
1.

Reviewing Agency Factual Determinations.

Under the APA, a reviewing court must accept an agency’s factual determinations in
informal proceedings unless they are “arbitrary [or] capricious,” 5 U.S.C. § 706(2)(A), and its
factual determinations in formal proceedings unless they are “unsupported by substantial
evidence,” 5 U.S.C. § 706(2)(E). The APA’s two linguistic formulations amount to a single
substantive standard of review. Ass’n of Data Processing Serv. Orgs., Inc. v. Bd. of Govs. of the
Fed. Reserve Sys., 745 F.2d 677, 683-84 (D.C. Cir. 1984)(Scalia, J.)(explaining that, as to
factual findings, “there is no substantive difference between what [the arbitrary or capricious
standard] requires and what would be required by the substantial evidence test, since it is
impossible to conceive of a ‘nonarbitrary’ factual judgment supported only by evidence that is
not substantial in the APA sense” (emphasis in original)). See also id. at 684 (“[T]his does not
consign paragraph (E) of the APA’s judicial review section to pointlessness. The distinctive
function of paragraph (E) -- what it achieves that paragraph (A) does not -- is to require
substantial evidence to be found within the record of closed-record proceedings to which it

- 44 -

exclusively applies.” (emphasis in original)).
In reviewing agency action under the arbitrary-or-capricious standard, a court considers
the administrative record -- or at least those portions of the record that the parties provide -- and
not materials outside of the record. See 5 U.S.C. § 706(“In making the foregoing determinations,
the court shall review the whole record or those parts of it cited by a party.”); Fed. R. App. P. 16
(“The record on review or enforcement of an agency order consists of . . . the order
involved; . . . any findings or report on which it is based; and . . . the pleadings, evidence, and
other parts of the proceedings before the agency.”); Ass’n of Data Processing Serv. Orgs., Inc. v.
Bd. of Govs. of the Fed. Reserve Sys., 745 F.2d at 684 (“[W]hether the administrator was
arbitrary must be determined on the basis of what he had before him when he acted.”). See also
Franklin Sav. Ass’n v. Dir., Office of Thrift Supervision, 934 F.2d 1127, 1137 (10th
Cir. 1991)(“[W]here Congress has provided for judicial review without setting forth . . .
procedures to be followed in conducting that review, the Supreme Court has advised such review
shall be confined to the administrative record and, in most cases, no de novo proceedings may be
had.”). Tenth Circuit precedent indicates, however, that the ordinary evidentiary rules regarding
judicial notice apply when a court reviews agency action. See New Mexico ex. rel. Richardson
v. Bureau of Land Mgmt., 565 F.3d 683, 702 n.21 (10th Cir. 2009)(citing Fed. R. Evid.
201(b))(“We take judicial notice of this document, which is included in the record before us in
[another case].”); id. at 702 n.22 (“We conclude that the occurrence of Falcon releases is not
subject to reasonable factual dispute and is capable of determination using sources whose
accuracy cannot reasonably be questioned, and we take judicial notice thereof.”). In contrast, the
United States Courts of Appeals for the Ninth and Eleventh Circuits have held that taking
judicial notice is inappropriate in APA reviews absent extraordinary circumstances or

- 45 -

inadvertent omission from the administrative record. See Compassion Over Killing v. U.S. Food
& Drug Administration, 849 F.3d 849, 852 n.1 (9th Cir. 2017); National Min. Ass’n v. Secretary
U.S. Dep’t of Labor, 812 F.3d 843, 875 (11th Cir. 2016).
To fulfill its function under the APA, a reviewing court should engage in a “thorough,
probing, in-depth review” of the record before it when determining whether an agency’s decision
survives arbitrary-or-capricious review. Wyoming v. United States, 279 F.3d 1214, 1238 (10th
Cir. 2002)(citation omitted). The Tenth Circuit explains:
In determining whether the agency acted in an arbitrary and capricious manner,
we must ensure that the agency decision was based on a consideration of the
relevant factors and examine whether there has been a clear error of judgment.
We consider an agency decision arbitrary and capricious if the agency relied on
factors which Congress had not intended it to consider, entirely failed to consider
an important aspect of the problem, offered an explanation for its decision that
runs counter to the evidence before the agency, or is so implausible that it could
not be ascribed to a difference in view or the product of agency expertise.
Colo. Envtl. Coal. v. Dombeck, 185 F.3d 1162, 1167 (10th Cir. 1999). Arbitrary-or-capricious
review requires a district court “to engage in a substantive review of the record to determine if
the agency considered relevant factors and articulated a reasoned basis for its conclusions,”
Olenhouse, 42 F.3d at 1580, but it is not to assess the wisdom or merits of the agency’s decision,
see Colo. Envtl. Coal. v. Dombeck, 185 F.3d at 1172. The agency must articulate the same
rationale for its findings and conclusions on appeal upon which it relied in its internal
proceedings. See SEC v. Chenery Corp., 318 U.S. 80 (1943). While the court may not supply a
reasoned basis for the agency’s action that the agency does not give itself, the court should
“uphold a decision of less than ideal clarity if the agency’s path may reasonably be discerned.”
Bowman Transp., Inc. v. Arkansas-Best Freight Sys., Inc., 419 U.S. 281, 286 (1974)(internal
citations omitted).

- 46 -

2.

Reviewing Agency Legal Interpretations.

In promulgating and enforcing regulations, agencies must interpret federal statutes, their
own regulations, and the Constitution of the United States of America, and Courts reviewing
those interpretations apply three different deference standards, depending on the kind of law at
issue. First, the federal judiciary accords considerable deference to an agency’s interpretation of
a statute that Congress has tasked it with enforcing.

See United States v. Undetermined

Quantities of Bottles of an Article of Veterinary Drug, 22 F.3d 235, 238 (10th Cir. 1994). This is
known as Chevron deference, named after the supposedly seminal case, Chevron, U.S.A., Inc. v.
Natural Resource Defense Council, Inc., 467 U.S. 837 (1984)(“Chevron”). 9 Chevron deference
is a two-step process10 that first asks whether the statutory provision in question is clear and, if it
is not, then asks whether the agency’s interpretation of the unclear statute is reasonable. As the
Tenth Circuit has explained,
9

The case itself is unremarkable, uninstructive, does not explicitly outline the nowfamiliar two-step process of applying Chevron deference, and does not appear to have been
intended to become a “big name” case at all. Its author, the Honorable John Paul Stevens,
former Associate Justice of the Supreme Court, insists that the case was never intended to create
a regime of deference, and, in fact, Justice Stevens became one of Chevron deference’s greatest
detractors in subsequent years. See generally Charles Evans Hughes, Justice Stevens and the
Chevron Puzzle, 106 Nw. U. L. Rev. 551 (2012).
10

There is, additionally, a threshold step -- the so-called step zero -- which asks whether
Chevron deference applies to the agency decision at all. See Cass R. Sunstein, Chrevron Step
Zero, 92 Va. L. Rev. 187 (2006). Step zero asks: (i) whether the agency is Chevron-qualified,
meaning whether the agency involved is the agency charged with administering the statute -- for
example, the EPA administers a number of statutes, among them the Clean Air Act, Pub. L. No.
88-206, 77 Stat. 392; (ii) whether the decision fits within the category of interpretations afforded
the deference -- interpretation of contracts, the Constitution, and the agency’s own regulations
are not afforded Chevron deference, see, e.g., U.S. West, Inc. v. FCC, 182 F.3d 1224 (10th
Cir. 1999)(“[A]n unconstitutional interpretation is not entitled to Chevron deference.”); and
(iii) whether Congress intended the agency to “speak with the force of law” in making the
decision in question, United States v. Mead Corp., 533 U.S. 218, 229 (2001) -- opinion letters by
the agency, for example, do not speak with the force of law and are thus not entitled to Chevron
deference, see Christensen v. Harris Cty., 529 U.S. 576 (2000). An affirmative answer to all
three inquiries results in the agency’s decision passing step zero.
- 47 -

we must be guided by the directives regarding judicial review of administrative
agency interpretations of their organic statutes laid down by the Supreme Court in
Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837
. . . (1984). Those directives require that we first determine whether Congress has
directly spoken to the precise question at issue. If the congressional intent is
clear, we must give effect to that intent. If the statute is silent or ambiguous on
that specific issue, we must determine whether the agency’s answer is based on a
permissible construction of the statute.
United States v. Undetermined Quantities of Bottles of an Article of Veterinary Drug, 22 F.3d at
238 (citation omitted).
Chevron’s second step is all but toothless, because if the agency’s decision makes it to
step two, it is upheld almost without exception.

See Ronald M. Levin, The Anatomy of

Chevron: Step Two Reconsidered, 72 Chi.-Kent L. Rev. 1253, 1261 (1997)(“[T]he Court has
never once struck down an agency’s interpretation by relying squarely on the second Chevron
step.”

(footnote omitted)); Jason J. Czarnezki, An Empirical Investigation of Judicial

Decisionmaking, Statutory Interpretation, and the Chevron Doctrine in Environmental Law, 79
U. Colo. L. Rev. 767, 775 (2008)(“Due to the difficulty in defining step two, courts rarely strike
down agency action under step two, and the Supreme Court has done so arguably only twice.”).
Courts essentially never conclude that an agency’s interpretation of an unclear statute is
unreasonable.
Chevron’s first step, in contrast, has bite, but there is substantial disagreement about what
it means. In an earlier case, the Court noted the varying approaches that different Supreme Court
Justices have taken in applying Chevron deference:
The Court notices a parallel between the doctrine of constitutional
avoidance and the Chevron doctrine. Those Justices, such as Justice Scalia, who
are most loyal to the doctrines and the most likely to apply them, are also the most
likely to keep the “steps” of the doctrines separate: first, determining whether the
statute is ambiguous; and, only then, assessing the merits of various permissible
interpretations from the first step. These Justices are also the most likely to find
that the statute is unambiguous, thus obviating the need to apply the second step
of each doctrine. Those Justices more likely to find ambiguity in statutes are
- 48 -

more likely to eschew applying the doctrines in the first place, out of their distaste
for their second steps -- showing heavy deference to agencies for Chevron
doctrine, and upholding facially overbroad statutes, for constitutional avoidance.
Griffin v. Bryant, 30 F. Supp. 3d 1139, 1193 n.23 (D.N.M.2014)(Browning, J.). A number of
policy considerations animate Chevron deference, among them: (i) statutory interpretation, i.e.,
that Congress, by passing extremely open-ended and vague organic statutes, grants discretionary
power to the agencies to fill in the statutory gaps; (ii) institutional competency, i.e., that agencies
are more competent than the courts at filling out the substantive law in their field; (iii) political
accountability, i.e., that agencies, as executive bodies ultimately headed by the President of the
United States of America, can be held politically accountable for their interpretations; and
(iv) efficiency, i.e., that numerous, subject-matter specialized agencies can more efficiently
promulgate the massive amount of interpretation required to maintain the modern regulatory
state -- found in the Code of Federal Regulations and other places -- than a unified but Circuitfragmented federal judiciary can.
Second, when agencies interpret their own regulations -- to, for example, adjudicate
whether a regulated party was in compliance with them -- courts accord agencies what is known
as Auer or Seminole Rock deference. See Auer v. Robbins, 519 U.S. 452 (1997)(“Auer”);
Bowles v. Seminole Rock & Sand Co., 325 U.S. 410 (1945). This deference is applied in the
same manner as Chevron deference and is substantively identical. There would be little reason
to have a separate name for this doctrine, except that its logical underpinnings are much shakier,
and its future is, accordingly, more uncertain. Justice Scalia, after years of applying the doctrine
followed by years of questioning its soundness, finally denounced Auer deference in 2013 in his
dissent in Decker v. Northwest Environmental Defense Center, 568 U.S. 597 (2013). The Court
cannot describe the reasons for Justice Scalia’s abandonment of the doctrine better than the
Justice himself:
- 49 -

For decades, and for no good reason, we have been giving agencies the
authority to say what their rules mean, under the harmless-sounding banner of
“defer[ring] to an agency’s interpretation of its own regulations.” Talk America,
Inc. v. Michigan Bell Telephone Co., [564] U.S. [50], 131 S. Ct. 2254, 2265, 180
L.Ed.2d 96 (2011) (Scalia, J., concurring). This is generally called Seminole Rock
or Auer deference.
....
The canonical formulation of Auer deference is that we will enforce an
agency’s interpretation of its own rules unless that interpretation is “plainly
erroneous or inconsistent with the regulation.” But of course whenever the
agency’s interpretation of the regulation is different from the fairest reading, it is
in that sense “inconsistent” with the regulation. Obviously, that is not enough, or
there would be nothing for Auer to do. In practice, Auer deference is Chevron
deference applied to regulations rather than statutes. The agency’s interpretation
will be accepted if, though not the fairest reading of the regulation, it is a
plausible reading -- within the scope of the ambiguity that the regulation contains.
Our cases have not put forward a persuasive justification for Auer
deference. The first case to apply it, Seminole Rock, offered no justification
whatever -- just the ipse dixit that “the administrative interpretation . . . becomes
of controlling weight unless it is plainly erroneous or inconsistent with the
regulation.” Our later cases provide two principal explanations, neither of which
has much to be said for it. First, some cases say that the agency, as the drafter of
the rule, will have some special insight into its intent when enacting it. The
implied premise of this argument -- that what we are looking for is the agency’s
intent in adopting the rule -- is false. There is true of regulations what is true of
statutes. As Justice Holmes put it: “[w]e do not inquire what the legislature
meant; we ask only what the statute means.” Whether governing rules are made
by the national legislature or an administrative agency, we are bound by what they
say, not by the unexpressed intention of those who made them.
The other rationale our cases provide is that the agency possesses special
expertise in administering its “‘complex and highly technical regulatory
program.’” That is true enough, and it leads to the conclusion that agencies and
not courts should make regulations. But it has nothing to do with who should
interpret regulations -- unless one believes that the purpose of interpretation is to
make the regulatory program work in a fashion that the current leadership of the
agency deems effective. Making regulatory programs effective is the purpose of
rulemaking, in which the agency uses its “special expertise” to formulate the best
rule. But the purpose of interpretation is to determine the fair meaning of the
rule -- to “say what the law is.” Not to make policy, but to determine what policy
has been made and promulgated by the agency, to which the public owes
obedience. Indeed, since the leadership of agencies (and hence the policy
preferences of agencies) changes with Presidential administrations, an agency
head can only be sure that the application of his “special expertise” to the issue
addressed by a regulation will be given effect if we adhere to predictable
principles of textual interpretation rather than defer to the “special expertise” of
his successors. If we take agency enactments as written, the Executive has a

- 50 -

stable background against which to write its rules and achieve the policy ends it
thinks best.
Another conceivable justification for Auer deference, though not one that
is to be found in our cases, is this: If it is reasonable to defer to agencies regarding
the meaning of statutes that Congress enacted, as we do per Chevron, it is a
fortiori reasonable to defer to them regarding the meaning of regulations that they
themselves crafted. To give an agency less control over the meaning of its own
regulations than it has over the meaning of a congressionally enacted statute
seems quite odd.
But it is not odd at all. The theory of Chevron (take it or leave it) is that
when Congress gives an agency authority to administer a statute, including
authority to issue interpretive regulations, it implicitly accords the agency a
degree of discretion, which the courts must respect, regarding the meaning of the
statute. While the implication of an agency power to clarify the statute is
reasonable enough, there is surely no congressional implication that the agency
can resolve ambiguities in its own regulations. For that would violate a
fundamental principle of separation of powers -- that the power to write a law and
the power to interpret it cannot rest in the same hands. “When the legislative and
executive powers are united in the same person . . . there can be no liberty;
because apprehensions may arise, lest the same monarch or senate should enact
tyrannical laws, to execute them in a tyrannical manner.” Montesquieu, Spirit of
the Laws bk. XI, at 151-152 (O. Piest ed., T. Nugent transl. 1949). Congress
cannot enlarge its own power through Chevron -- whatever it leaves vague in the
statute will be worked out by someone else. Chevron represents a presumption
about who, as between the Executive and the Judiciary, that someone else will
be. (The Executive, by the way -- the competing political branch -- is the less
congenial repository of the power as far as Congress is concerned.) So
Congress’s incentive is to speak as clearly as possible on the matters it regards as
important.
But when an agency interprets its own rules -- that is something else.
Then the power to prescribe is augmented by the power to interpret; and the
incentive is to speak vaguely and broadly, so as to retain a “flexibility” that will
enable “clarification” with retroactive effect. “It is perfectly understandable” for
an agency to “issue vague regulations” if doing so will “maximiz[e] agency
power.” Combining the power to prescribe with the power to interpret is not a
new evil: Blackstone condemned the practice of resolving doubts about “the
construction of the Roman laws” by “stat[ing] the case to the emperor in writing,
and tak[ing] his opinion upon it.” 1 Wm. Blackstone, Commentaries on the Laws
of England 58 (1765). And our Constitution did not mirror the British practice of
using the House of Lords as a court of last resort, due in part to the fear that he
who has “agency in passing bad laws” might operate in the “same spirit” in their
interpretation. The Federalist No. 81, at 543-544 (Alexander Hamilton)(J. Cooke
ed. 1961). Auer deference encourages agencies to be “vague in framing
regulations, with the plan of issuing ‘interpretations’ to create the intended new
law without observance of notice and comment procedures.” Auer is not a logical
corollary to Chevron but a dangerous permission slip for the arrogation of power.

- 51 -

It is true enough that Auer deference has the same beneficial pragmatic
effect as Chevron deference: The country need not endure the uncertainty
produced by divergent views of numerous district courts and courts of appeals as
to what is the fairest reading of the regulation, until a definitive answer is finally
provided, years later, by this Court. The agency’s view can be relied upon, unless
it is, so to speak, beyond the pale. But the duration of the uncertainty produced
by a vague regulation need not be as long as the uncertainty produced by a vague
statute. For as soon as an interpretation uncongenial to the agency is pronounced
by a district court, the agency can begin the process of amending the regulation to
make its meaning entirely clear. The circumstances of this case demonstrate the
point. While these cases were being briefed before us, EPA issued a rule
designed to respond to the Court of Appeals judgment we are reviewing. It did so
(by the standards of such things) relatively quickly: The decision below was
handed down in May 2011, and in December 2012 the EPA published an
amended rule setting forth in unmistakable terms the position it argues here. And
there is another respect in which a lack of Chevron-type deference has less severe
pragmatic consequences for rules than for statutes. In many cases, when an
agency believes that its rule permits conduct that the text arguably forbids, it can
simply exercise its discretion not to prosecute. That is not possible, of course,
when, as here, a party harmed by the violation has standing to compel
enforcement.
In any case, however great may be the efficiency gains derived from Auer
deference, beneficial effect cannot justify a rule that not only has no principled
basis but contravenes one of the great rules of separation of powers: He who
writes a law must not adjudge its violation.
Decker v. Nw. Envtl. Def. Ctr., 568 U.S. 597, 616-21 (Scalia, J., dissenting)(alterations in
original)(citations omitted). Although the Court shares Justice Scalia’s concerns about Auer
deference, it is, for the time being, the law of the land, and, as a federal district court, the Court
must apply it.
Last, courts afford agencies no deference in interpreting the Constitution. See U.S. West,
Inc. v. FCC, 182 F.3d 1224, 1231 (10th Cir. 1999)(“[A]n unconstitutional interpretation is not
entitled to Chevron deference. . . . [D]eference to an agency interpretation is inappropriate not
only when it is conclusively unconstitutional, but also when it raises serious constitutional
questions.” (citing, e.g., Rust v. Sullivan, 500 U.S. 173, 190-91 (1991))). Courts have superior
competence in interpreting -- and constitutionally vested authority and responsibility to
interpret -- the Constitution’s content. The presence of a constitutional claim does not take a
- 52 -

court’s review outside of the APA, however -- § 706(2)(B) specifically contemplates
adjudication of constitutional issues -- and courts must still respect agency fact-finding and the
administrative record when reviewing agency action for constitutional infirmities; they just
should not defer to the agency on issues of substantive legal interpretation. See, e.g., Robbins v.
U.S. Bureau of Land Mgmt., 438 F.3d 1074, 1085 (10th Cir. 2006)(“We review Robbins’
[constitutional] due process claim against the [agency] under the framework set forth in the
APA.”).
3.

Waiving Sovereign Immunity.

The APA waives sovereign immunity with respect to non-monetary claims. See 5 U.S.C.
§ 702. The statute provides:
An action in a court of the United States seeking relief other than money damages
and stating a claim that an agency or an officer or employee thereof acted or
failed to act in an official capacity or under color of legal authority shall not be
dismissed nor relief therein be denied on the ground that it is against the United
States or that the United States is an indispensable party. The United States may
be named as a defendant in any such action, and a judgment or decree may be
entered against the United States:
5 U.S.C. § 702. Claims for money damages seek monetary relief “to substitute for a suffered
loss.” Normandy Apartments, Ltd. v. U.S. Dep’t of Hous. & Urban Dev., 554 F.3d 1290, 1298
(10th Cir. 2009)(emphasis in original). Claims that do not seek monetary relief or that seek
“specific remedies that have the effect of compelling monetary relief” are not claims for
monetary damages. Normandy Apartments, Ltd. v. U.S. Dep’t of Hous. & Urban Dev., 554 F.3d
at 1298. To determine whether a claim seeks monetary relief, a court must “look beyond the
face of the complaint” and asses the plaintiff’s prime object or essential purpose; “‘[a] plaintiff’s
prime objective or essential purpose is monetary unless the non-monetary relief sought has
significant prospective effect or considerable value apart from the claim for monetary relief.’”
Normandy Apartments, Ltd. v. U.S. Dep’t of Hous. & Urban Dev., 554 F.3d at 1296 (quoting
- 53 -

Burkins v. United States, 112 F.3d 444, 449 (10th Cir. 1997)).
The APA’s sovereign immunity waiver for claims “seeking relief other than money
damages” does not apply, however, “if any other statute that grants consent to suit expressly or
impliedly forbids the relief which is sought.” 5 U.S.C. § 702. The Tucker Act, 28 U.S.C.
§§ 1346, 1491, permits district courts to hear some claims against the United States, but it also
states that “district courts shall not have jurisdiction of any civil action or claim against the
United States founded upon any express or implied contract with the United States.” 28 U.S.C.
§ 1346(a)(2). It follows that the APA does not waive the United States’ sovereign immunity as
to contract claims even when those claims seek relief other than money damages, such as
declaratory or injunctive relief. See Normandy Apartments, Ltd. v. U.S. Dep’t of Hous. & Urban
Dev., 554 F.3d at 1295. Consequently, two questions determine whether the APA waives the
United States’ sovereign immunity as to a particular claim: “First, does [the] claim seek ‘relief
other than money damages,’ such that the APA’s general waiver of sovereign immunity is even
implicated? Second, does the Tucker Act expressly or impliedly forbid the relief that Normandy
seeks, such that the APA’s waiver does not apply?” Normandy Apartments, Ltd. v. U.S. Dep’t
of Hous. & Urban Dev., 554 F.3d at 1296 (quoting 5 U.S.C. § 702).
ANALYSIS
The Court concludes that: (i) the APA waives the United States’ sovereign immunity for
Health Connections’ claims; (ii) incorporating statewide average premiums in HHS’ riskadjustment formula is not contrary to law, but it is arbitrary and capricious, because the
administrative record indicates that HHS assumed, erroneously, that the ACA requires risk
adjustment to be budget neutral, and all of HHS’ reasons for using the statewide average
premium rely on that budget neutrality assumption; (iii) HHS’ methods of predicting healthcare

- 54 -

costs for HCC and non-HCC patients is not arbitrary and capricious; (iv) HHS’ risk-adjustment
practices regarding partial-year enrollees and prescription-drug data are not arbitrary and
capricious; and (v) HHS’ risk-adjustment formula does not effectively eliminate bronze-level
health-insurance plans. The Court, accordingly, sets aside and vacates HHS’ action as to the
statewide average premium rules and remands the case to the agency for further proceedings.
I.

THE APA WAIVES THE UNITED STATES’ SOVEREIGN IMMUNITY FOR
HEALTH CONNECTIONS’ CLAIMS.
The APA waives sovereign immunity with respect to non-monetary claims. See 5 U.S.C.

§ 702. The statute provides:
An action in a court of the United States seeking relief other than money damages
and stating a claim that an agency or an officer or employee thereof acted or
failed to act in an official capacity or under color of legal authority shall not be
dismissed nor relief therein be denied on the ground that it is against the United
States or that the United States is an indispensable party. The United States may
be named as a defendant in any such action, and a judgment or decree may be
entered against the United States:
5 U.S.C. § 702. Claims for money damages seek monetary relief “to substitute for a suffered
loss.” Normandy Apartments, Ltd. v. U.S. Dep’t of Hous. & Urban Dev., 554 F.3d at 1298
(emphasis in original). Claims that do not seek monetary relief or that seek “specific remedies
that have the effect of compelling monetary relief” are not claims for monetary damages.
Normandy Apartments, Ltd. v. U.S. Dep’t of Hous. & Urban Dev., 554 F.3d at 1298. To
determine whether a claim seeks monetary relief, a court must “look beyond the face of the
complaint” and asses the plaintiff’s prime object or essential purpose; “‘[a] plaintiff’s prime
objective or essential purpose is monetary unless the non-monetary relief sought has significant
prospective effect or considerable value apart from the claim for monetary relief.’” Normandy
Apartments, Ltd. v. U.S. Dep’t of Hous. & Urban Dev., 554 F.3d at 1296 (quoting Burkins v.
United States, 112 F.3d 444, 449 (10th Cir. 1997)).

- 55 -

The APA’s sovereign immunity waiver for claims “seeking relief other than money
damages” does not apply, however, “if any other statute that grants consent to suit expressly or
impliedly forbids the relief which is sought.” 5 U.S.C. § 702. The Tucker Act, 28 U.S.C.
§§ 1346, 1491, permits district courts to hear some claims against the United States, but it also
states that “district courts shall not have jurisdiction of any civil action or claim against the
United States founded upon any express or implied contract with the United States.” 28 U.S.C.
§ 1346(a)(2). It follows that the APA does not waive the United States’ sovereign immunity as
to contract claims even when those claims seek relief other than money damages, such as
declaratory or injunctive relief. See Normandy Apartments, Ltd. v. U.S. Dep’t of Hous. & Urban
Dev., 554 F.3d at 1295. Consequently, two questions determine whether the APA waives the
United States’ sovereign immunity as to a particular claim: “First, does [the] claim seek ‘relief
other than money damages,’ such that the APA’s general waiver of sovereign immunity is even
implicated? Second, does the Tucker Act expressly or impliedly forbid the relief that Normandy
seeks, such that the APA’s waiver does not apply?” Normandy Apartments, Ltd. v. U.S. Dep’t
of Hous. & Urban Dev., 554 F.3d at 1296 (quoting 5 U.S.C. § 702).
The Court concludes that Health Connections’ claims are not for monetary relief. Health
Connections asks the Court to declare HHS’ risk adjustment methodology invalid, declare that
HHS’ risk adjustment methodology must be revised, and enjoin further application of that
methodology. See Complaint at 45. Health Connections’ claims for declaratory and injunctive
relief could cause HHS to pay money to Health Connections, because “[i]f this Court grants the
requested relief, HHS will then be directed to conduct a new rulemaking process.” Plaintiff’s
Reply at 29. “If it is determined under that new scheme that HHS owes money back to NMHC
. . . HHS can issue that refund or NMHC can sue in the court of Federal Claims.” Plaintiff’s

- 56 -

Reply at 29. That granting relief to Health Connections “might enable a subsequent claim for
monetary relief . . . does not preclude the district court from exercising jurisdiction at this point.”
Normandy Apartments, Ltd. v. U.S. Dep’t of Hous. & Urban Dev., 554 F.3d at 1298. Health
Connections’ challenges to HHS’ risk adjustment methodology would, if successful, alter Health
Connections’ risk adjustment payments for future years. Consequently, Health Connections
seeks relief with significant prospective value -- and not monetary relief -- even though
providing relief to Health Connections could cause HHS to refund a portion of the risk
adjustment payments that Health Connections has already made. See Normandy Apartments,
Ltd. v. U.S. Dep’t of Hous. & Urban Dev., 554 F.3d at 1298 (concluding that a plaintiff’s claim
retained significant prospective value such that its “primary object remains securing equitable
relief” even though the claim’s success, “over a year after HUD ceased disbursing funds to the
company” could cause the agency to make belated payments).
The Court determines that, even if Health Connections’ challenge to HHS’ risk
adjustment methodology seek monetary relief, it does not seek money damages. See Normandy
Apartments, Ltd. v. U.S. Dep’t of Hous. & Urban Dev., 554 F.3d at 1296. The Supreme Court
has explained that not all judicial remedies requiring monetary payments are characterized as
damages. See Bowen v. Mass., 487 U.S. at 893-94. “The fact that a judicial remedy may require
one party to pay money to another is not a sufficient reason to characterize the relief as ‘money
damages.’”

Bowen v. Mass., 487 U.S. at 893-94.

For example, the Supreme Court has

“recognized that relief that orders a town to reimburse parents for educational costs that
Congress intended the town to pay is not ‘damages.’” Bowen v. Mass., 487 U.S. at 893. In
contrast to damages, “‘[r]eimbursement merely requires the Town to belatedly pay expenses that
it should have paid all along.’” Bowen v. Mass., 487 U.S. at 894 (quoting School Committee of

- 57 -

Burlington v. Department of Education of Mass., 471 U.S. 359, 370-71 (1985)). The Supreme
Court also quoted the Honorable Robert Bork, United States Circuit Judge, who explained that
“‘[d]amages are given to the plaintiff to substitute for a suffered loss, whereas specific remedies
are not substitute remedies at all, but attempt to give the plaintiff the very thing to which he was
entitled.’” Bowen v. Mass., 487 U.S. at 895 (quoting Maryland Dept. of Human Resources v.
Department of Health and Human Services, 763 F.2d 1441, 1446 (D.C. Cir. 1985)(Bork,
J.)(emphasis in Maryland Dept. of Human Resources v. Department of Health and Human
Services)).
Here, Health Connections ultimately wants the Court “to invalidate regulations for past
years so that HHS can then fix the regulatory scheme.” Plaintiff’s Reply at 29. “If it is
determined that under that new scheme that HHS owes money back to NMHC, HHS can issue
that refund or NMHC can sue in the court of federal claims.” Plaintiff’s Reply at 29 (emphasis
added). These statements indicate that Health Connections is not seeking damages, because
“ow[ing] money back” or requesting a “refund” is not a “‘substitute for a suffered loss.’” Bowen
v. Mass., 487 U.S. at 895 (quoting Maryland Dept. of Human Resources v. Department of Health
and Human Services, 763 F.2d at 1446 (emphasis in Maryland Dept. of Human Resources v.
Department of Health and Human Services)). Rather, Health Connections requests “‘the very
thing to which [Health Connections is] entitled’” or believes that it is entitled. Bowen v. Mass.,
487 U.S. at 895 (quoting Maryland Dept. of Human Resources v. Department of Health and
Human Services, 763 F.2d at 1446 (alteration added)). The Court therefore concludes that
Health Connections is not requesting damages.
The Court also determines that Health Connections’ claim is founded on the ACA and
HHS regulations and not on a contract with the federal government, so the Tucker act does not

- 58 -

expressly or implicitly bar that claim. See Normandy Apartments, Ltd. v. U.S. Dep’t of Hous. &
Urban Dev., 554 F.3d at 1299 (“Therefore, in order to determine whether the Tucker Act
impliedly forbids the relief sought in this case, we must evaluate whether Normandy's claim is
properly understood as one founded on contract or on the federal Constitution, statutes, or
regulations.”).

Specifically, Health Connections’ claims relate to HHS’ risk adjustment

formula’s use of state average premiums. See Plaintiff Mem. at 24-25 (citing 2014 Final Rule,
78 Fed. Reg. at 15,430-34); Plaintiff Mem. at 24 (citing 42 U.S.C. § 18063(a)(1)). Because
Health Connections’ claims arise under HHS regulations and under the ACA, and not under a
contract,11 the Court concludes that the Tucker Act does not forbid the relief sought. See
Normandy Apartments, Ltd. v. U.S. Dep’t of Hous. & Urban Dev., 554 F.3d at 1299.
Accordingly, the Court concludes that the APA waives the United States’ sovereign immunity as
to Health Connections’ claims.
II.

HHS’ USE OF STATEWIDE AVERAGE PREMIUMS IS NOT CONTRARY TO
LAW, BUT IS ARBITRARY AND CAPRICIOUS.
Health Connections argues both that HHS’ regulations violate the ACA, and that those

regulations are arbitrary and capricious. See Plaintiff Mem. at 16. The Court determines that the
HHS’ risk adjustment regulations are not contrary to law, because the ACA permits risk
11

In becoming a CO-OP, Health Connections signed a loan agreement with HHS to fund
Health Connections’ initial formation and operation in New Mexico. See Plaintiff Mem. ¶ 23, at
11 (citing Hickey Declaration at 5 (dated October 5, 2016)(NMHC000867)). This loan
agreement is not, however, the source of this litigation. Rather, Health Connections’ claims
relate to HHS’ risk adjustment formula’s use of state average premiums. See Plaintiff Mem. at
24-25 (citing 78 Fed. Reg. at 15,430-34); Plaintiff Mem. at 24 (citing 42 U.S.C. § 18063(a)(1)).
“When the source of rights asserted is constitutional, statutory, or regulatory in nature, the fact
that resolution of the claim requires some reference to contract does not magically transform
[the] action . . . into one on the contract and deprive the court of jurisdiction it might otherwise
have.” Normandy Apartments, Ltd. v. U.S. Dep’t of Hous. & Urban Dev., 554 F.3d at 1300
(quotations omitted). Because the source of rights asserted here are statutory and regulatory, the
Court concludes that the mere existence of this loan agreement does not deprive the Court of
jurisdiction.
- 59 -

adjustment payments to be based on factors other than actuarial risk and because, while the ACA
does not require risk adjustment to be budget neutral, it also does not forbid budget neutrality.
The Court also determines, however, that HHS’ risk adjustment regulations -- specifically their
use of the statewide average premium -- are arbitrary and capricious, because HHS’ justifications
for using the statewide average premium instead of a plan’s own premium all assume that the
ACA requires risk adjustment to be budget neutral, which is not correct. Nevertheless, HHS
could have justified its promulgation of budget neutral regulations if it determined that budget
neutrality was a worthy policy goal. HHS never made such a determination in the record,
however, and the Court considers only the reasons that the agency actually gave and not the
reasons that the agency might have given when determining agency action was arbitrary and
capricious.
A.

HHS’ RISK ADJUSTMENT REGULATIONS ARE NOT CONTRARY TO
LAW.

Health Connections makes two contentions regarding the ACA’s risk adjustment
provisions: (i) those provisions “mandate that risk adjustment assessments be based solely upon
actuarial risk,” Plaintiff Mem. at 17; and (ii) “there is no statutory requirement that risk
adjustment be budget neutral,” Plaintiff Mem. at 22. To evaluate those contentions, the Court
must determine whether, under Chevron, it must defer to HHS’ contrary conclusions about the
statute, i.e., that the ACA does not mandate that risk adjustment payments be based solely on
actuarial risk and that the ACA requires risk adjustment to be budget neutral. Consequently, the
Court must determine whether the ACA’s risk adjustment provisions are ambiguous regarding
the permissible bases for risk adjustment payments and whether risk adjustment must be budget
neutral. See Chevron, 467 U.S. at 842-43. If the ACA is ambiguous on either point, the Court
must determine whether HHS’ ACA interpretation is reasonable. See Chevron, 467 U.S. at 843

- 60 -

(“[T]he question for the court is whether the agency’s answer is based on a permissible
construction of the statute.”).
The ACA does not clearly require risk adjustment payments to be based solely on
actuarial risk. That the ACA commands “[t]he Secretary, in consultation with States, [to]
establish criteria and methods to be used in carrying out the risk adjustment activities” indicates
that the ACA does not oblige HHS to use actuarial risk as the sole actuarial risk criterion. 42
U.S.C. § 18063(b). Telling HHS to establish risk adjustment criteria would make no sense
otherwise. The ACA’s language stating that health plans and health insurance issuers must make
a risk adjustment payment if their enrollees have below-average actuarial risk, see 42 U.S.C.
§ 18063(a)(1), while health plans and health insurance issuers must receive a risk adjustment
payment if their enrollees have above-average actuarial risk, see 42 U.S.C. § 18063(a)(2), does
not mean that criteria other than actuarial risk cannot be used when determining the magnitude of
those payments. Health Connections insists that the statute is clear that, in calculating the
charge, the State or HHS must only consider actuarial risk as a factor, see Plaintiff Mem. at 17;
Plaintiff’s Reply at 11, but the Court sees no such requirement plainly in the statute’s text. The
statute does not say anything about how to calculate the charge; it says only “each state shall
assess a charge. . . .” 42 U.S.C. § 18063(a)(1). From the text, the charge assessed could be any
amount. While a potential -- perhaps even a plausible -- reading of the statutory text might be
that the assessment should be proportional to the relative disparity between the average actuarial
risk and the insurer’s actuarial risk, there is no such explicit or implicit requirement
That the ACA does not clearly require risk adjustment to be based solely on actuarial risk
means that HHS determination to the same effect is reasonable, although the Court does not,
technically, need to proceed to Chevron step two, because the statute is clear on this point. See

- 61 -

Chevron, 467 U.S. at 842-43 (“First, always, is the question whether Congress has directly
spoken to the precise question at issue. If the intent of Congress is clear, that is the end of the
matter.”). It also means that the Court is not persuaded by Health Connections’ argument that
HHS’ risk adjustment regulations are contrary to law insofar as they incorporate criteria other
than actuarial risk when calculating risk adjustment payments.
The ACA, however, does not unambiguously require the risk adjustment program to be
budget neutral. It is an established principle of statutory construction that, when “Congress
includes particular language in one section of a statute but omits it in another section of the same
Act, it is generally presumed that Congress acts intentionally and purposely in the disparate
inclusion or exclusion.” Bates v. United States, 523 U.S. 23, 29-30 (1997). This canon of
construction applies when a court analyzes whether a statute is ambiguous for Chevron purposes.
See New Mexico v. Dep’t of Interior, 854 F.3d 1207, 1223 (10th Cir. 2017)(“In conducting our
Chevron step-one analysis, we [e]mploy [] traditional tools of statutory construction. These tools
include examination of the statute’s text, structure, purpose, history, and relationship to other
statutes.” (alterations in original)).
The ACA’s risk adjustment provisions include no explicit requirements regarding risk
adjustment’s budgetary implications or the lack thereof. See 42 U.S.C. § 18063. The ACA’s
reinsurance provisions, however, require reinsurance entities to fund their payments to health
insurance issuers who cover high risk individuals with the payments that those reinsurance
entities receive from health insurance issuers, i.e., the ACA specifies that outgoing reinsurance
payments equal incoming reinsurance payments.

See 42 U.S.C. § 18061(b)(1)(B).

Risk

adjustment and reinsurance serve similar functions, although risk adjustment is permanent while
reinsurance is temporary. Compare 2014 Final Rule, 78 Fed. Reg. at 15,411 (dated March 11,

- 62 -

2013)(A.R.000228)(“The reinsurance program will reduce the uncertainty of insurance risk in
the individual market by partially offsetting issuers’ risk associated with high-cost enrollees.”);
id. (“The transitional reinsurance program and the temporary risk corridors program, which
begin in 2014, are designed to provide issuers with greater payment stability as insurance market
reforms are implemented and Exchanges facilitate increased enrollment.”), with id. (“On an
ongoing basis, the risk adjustment program is intended to provide increased payments to health
insurance issuers that attract higher-risk populations, such as those with chronic conditions, and
reduce the incentives for issuers to avoid higher-risk enrollees.”). That the ACA contains an
explicit provision stating that incoming reinsurance payments fund outgoing reinsurance
payments and omits any such language regarding risk adjustment payments means that outgoing
risk adjustment payments need not equal incoming risk adjustment payments. Because the
statute is clear that budget neutrality for the risk adjustment regulations is not required, the Court
need not accept HHS’ contention that the ACA requires budget neutrality on those regulations.
See National Cable & Telecommunications Ass’n v. Brand X Internet Servs., 545 U.S. 967, 985
(2005)(noting that, when a “statute unambiguously requires the court’s construction,” the court’s
interpretation trumps the agency’s); Chevron, 467 U.S. at 842-43.
Although HHS promulgated its risk adjustment regulations under the erroneous belief
that risk adjustment must be budget neutral, it does not follow, however, that those regulations
violate the APA. While nothing in the APA requires risk adjustment to be budget neutral,
nothing in the APA forbids budget neutrality either. See 42 U.S.C. § 18063. Accordingly, the
Court concludes that HHS’ risk adjustment regulations are not contrary to law.
B.

HHS’ RISK ADJUSTMENT REGULATIONS ARE ARBITRARY AND
CAPRICIOUS.

To determine whether HHS’ risk adjustment regulations survive arbitrary and capricious

- 63 -

review, the Court must determine whether HHS “examine[d] the relevant data and articulate[d] a
satisfactory explanation for its action including a rational connection between the facts found and
the choice made.” Encino Motorcars, LLC v. Navarro, 136 S. Ct. at 2125. “One of the basic
procedural requirements of administrative rulemaking is that an agency must give adequate
reasons for its decisions.” Encino Motorcars, LLC v. Navarro, 136 S. Ct. at 2125. The agency
must articulate the same rationale for its findings and conclusions on appeal upon which it relied
in its internal proceedings. See SEC v. Chenery Corp., 318 U.S. 80, 87 (1943)(“The grounds
upon which an administrative order must be judged are those upon which the record discloses
that its actions was based.”).
HHS published three different documents when first crafting its risk adjustment
methodology. First, on September 12, 2011, HHS -- specifically the Center for Consumer
Information and Insurance Oversight -- published a white paper titled “Risk Adjustment
Implementation Issues.” Risk Adjustment Implementation Issues at 1 (dated September 12,
2011)(A.R.004367)(“White Paper”). In the White Paper, HHS sets out a basic structure for
calculating risk adjustment transfers: “Payments and charges will be calculated by multiplying
plan risk relative to the market by a premium amount or ‘baseline premium.’ Further, they will
be calculated in a zero sum, budget-neutral manner.” White Paper at 13 (A.R.004379). HHS’
only explanation for its declaration that adjustment payments will be calculated budget neutrally
is that, “in contrast to some current risk adjustment methodologies, the Affordable Care Act’s
risk adjustment program is designed to be budget neutral.” White Paper at 4 (A.R.004370). In
short, HHS assumed that the ACA requires budget neutrality, and HHS does not give an
independent policy reason for requiring budget neutrality.
HHS continues by identifying two major issues when developing a methodology to

- 64 -

calculate risk adjustment transfers: (i) “how to establish the baseline premium”; and (ii) “how to
balance payments and charges.” White Paper at 13-14 (A.R.004379-80). HHS compares two
basic approaches to establishing the baseline premium, using an average premium or using a
plan’s own premium. See White Paper at 14 (A.R.004380). HHS reasons that using an average
premium will automatically balance payments and charges, because “the State average is a single
dollar amount for all plans, and plan risk scores average to 1.0, [so] the payments and charges are
equal in this approach.” White Paper at 15 (A.R.004381). HHS determines that using a plan’s
own premium will not necessarily produce equal payments and charges, so “[s]ince payment and
charge transfers will be budget neutral, a method is needed to balance them if payments are
greater than charges or vice versa.” White Paper at 15 (A.R.004381). Again, HHS gives no
policy reason for requiring budget neutrality. See White Paper at 15 (A.R.004381). Finally,
HHS also considers balancing risk adjustment payments and charges by increasing charges on a
pro-rated basis, decreasing payments on a pro-rated basis, or by doing both. See White Paper at
15 (A.R.004381).
In its 2014 Proposed Rule, HHS decides to use a state’s average premium as the baseline
premium when calculating risk adjustment charges and payments. See HHS Notice of Benefit
and Payment Parameters for 2014, 77 Fed. Reg. 73,118, 73,139 (dated December 7,
2012)(A.R.000134)(“2014 Proposed Rule”). HHS articulates the rationale for its decision to use
a state average premium and not a plan’s own premium:
The approaches that used plans’ own premiums resulted in unbalanced payment
transfers, requiring a balancing adjustment to yield transfers that net to zero.
These examples also demonstrated that the balancing adjustments could introduce
differences in premiums across plans that were not consistent with features of the
plan (for example, AV or differences in costs and utilization patterns across rating
areas). A balancing adjustment would likely vary from year to year, and could
add uncertainty to the rate development process (that is, plan actuaries would
need to factor the uncertainty of the balancing adjustment into their transfer

- 65 -

estimates).
2014 Proposed Rule, 77 Fed. Reg. at 73,139 (A.R.000134). According to HHS, “transfers net to
zero when the State average premium is used as the basis for calculating transfers,” so a
balancing adjustment is unnecessary, and “[t]he State average premium provides a
straightforward and predictable benchmark for estimating transfers.” 2014 Proposed Rule, 77
Fed. Reg. at 73,139 (A.R.000134). Thus, its reasoning is premised on budget neutrality and
predictability. The proposed rule does not, however, provide a policy rationale for budget
neutrality. HHS’ 2014 Final Rule likewise employs a state average premium. See 2014 Final
Rule, 78 Fed. Reg. at 15,431 (dated March 11, 2013)(A.R.000248). In that rule, HHS reports
that it “agree[s] with commenters that use of a plan’s own premium may cause unintended
distortions in transfers.”

2014 Final Rule, 78 Fed. Reg. at 15,432 (dated March 11,

2013)(A.R.000249). HHS and the commenters’ reference to “distortions” is left unexplained.
2014 Final Rule, 78 Fed. Reg. at 15,432 (A.R.000249).
In evaluating whether HHS’ interpretation is permissible, the Court must rely upon the
rationale the agency articulated in its internal proceedings and not upon post hoc reasoning. See
Biodiversity Conservation Alliance v. Jiron, 762 F.3d 1036, 1060 (10th Cir. 2014)(“We will not,
for example, accept appellate counsel’s post-hoc rationalizations for agency action -- we must
uphold the agency’s action if at all, on the basis articulated by the agency itself.”).

The

rationales HHS relies upon in its internal proceedings are budget neutrality and predictability.
See Defendant Mem. at 22-23 (arguing that budget neutrality and predictability are rationales
which HHS relies upon). As explained above, HHS assumes budget neutrality as a given,
because it believes, erroneously, that the ACA requires it, see 77 Fed. Reg. at 73,139
(A.R.000134)(“The approaches that used plans’ own premiums resulted in unbalanced payment
transfers, requiring a balancing adjustment to yield transfers that net to zero.” (emphasis
- 66 -

added)), and HHS articulates no independent policy reason for requiring budget neutrality. The
ACA does not, however, require the risk adjustment payments to be budget neutral. See supra at
63-64.
That HHS erroneously reads the ACA’s risk adjustment provisions to require risk
adjustment payments equal risk adjustment charges infects its analysis of the relative merits of
using a state’s average premium when calculating risk adjustment transfers instead of using a
plan’s own premium. Because risk adjustment does not need to be budget neutral, HHS’ risk
adjustment methodology could use a plan’s own premium instead of a state’s average premium
without imposing a balancing adjustment. Accordingly, the problems that HHS identifies with
imposing a balancing adjustment -- that a balancing transfer is unpredictable and could produce
unintended distortions -- do not justify HHS’ aversion to using a plan’s own premium to
calculate risk adjustment transfers. See 2014 Proposed Rule, 77 Fed. Reg. at 73,139 (asserting
that “balancing adjustments could introduce differences in premiums across plans that were not
consistent with features of the plan”); id. (“A balancing adjustment would likely vary from year
to year, and could add uncertainty to the rate development process (that is, plan actuaries would
need to factor the uncertainty of the balancing adjustment into their transfer estimates).”). See
also Encino Motorcars, LLC v. Navarro, 136 S. Ct. at 2125 (stating that an agency must
articulate a rational connection between the facts it found and the choice it made). Indeed,
absent a balancing adjustment, using a plan’s own premium without imposing a balancing
adjustment is more predictable for health insurance issuers than using a state’s average premium;
each issuer sets its own premiums, but a state’s average premium depends on the decisions of all
the health insurance issuers in a particular state. But see 2014 Proposed Rule, 77 Fed. Reg. at
73,139 (concluding that, in contrast to a risk adjustment methodology based on a plan’s own

- 67 -

premium and incorporating a balancing adjustment, that “[t]he State average premium provides a
straightforward and predictable benchmark for estimating transfers.”).
While the ACA does not require risk adjustment to be budget neutral, nothing in the
statute forbids budget neutrality and designing risk adjustment to be budget neutral may be a
reasonable policy choice. See Minuteman Health, Inc. v. U.S. Dep’t of Health and Human
Servs., __ F.3d __, 2018 WL 627381, at *20 (D. Mass. 2018)(Saylor, J.)(“Although the statute
does not require the program to be budget-neutral, it does not prohibit the program from being
budget-neutral, either. . . . The question then becomes whether HHS’s decision to attempt to
operate the risk-adjustment program in a budget-neutral way was unreasonable or arbitrary.”).
Indeed, there may be excellent policy reasons for making the risk adjustment plan budget neutral.
For example, HHS may not have the funding to make up the shortfall between the risk
adjustment charges and credits.12 Budget neutrality may also be a rational policy decision, so
that HHS may allocate discretionary funds to other programs that more desperately need that
funding. The problem with invoking those policy rationales here, however, is that HHS never
articulates any public policy decision to operate risk adjustment in a budget neutral way; HHS’
only decision is to comply with a supposed statutory requirement. See 2014 Proposed Rule, 77
Fed. Reg. at 73,139 (“The approaches that used plans’ own premiums resulted in unbalanced
payment transfers, requiring a balancing adjustment to yield transfers that net to zero.”
(emphasis added)); White Paper at 15 (reasoning that “[b]alancing is needed for all options to
establish a baseline premium, except for the State average,” because “payments and charges are
equal” in that approach). See also Defendant Mem. at 22 (“[B]ecause the risk adjustment
12

The Court notes that, since Congress passed the ACA, Congress has barred HHS from
using its annual appropriations to fund different risk program’s shortfalls. See Consolidated and
Further Continuing Appropriations Act, 2015, Pub. L. No. 113-235, §227; Plaintiff Mem. at 24,
n.7. Given that action, it may be rational for HHS to make risk adjustment budget neutral.
- 68 -

program is self-funded and budget-neutral, payments and charges must balance.”). That HHS, in
designing its risk adjustment methodology, never considered whether budget neutrality was
sound public policy means that HHS cannot now appeal to budget neutrality’s public policy
benefits to justify its decision. See Bowman Transp., Inc. v. Arkansas-Best Freight System, Inc.,
419 U.S. at 285-86 (stating that “we may not supply a reasoned basis for the agency’s action the
agency itself has not given”). That HHS can reasonably conclude that budget neutrality is a
worthy public policy goal does not permit the Court, in reviewing HHS’ decisionmaking, to act
as though HHS actually considered the issue and reached that conclusion.13
HHS argues, however, that it articulates two other reasons for adopting its rule. First, it
contends that HHS reasons that statewide premiums better calculate actuarial risk than an
issuer’s own premium. See Defendant Mem. at 20-21 (citing 2014 Final Rule, 78 Fed. Reg. at
15,432 (A.R.000249)). The citation it directs the Court to support its point, however, is HHS’
cryptic sentence that “[w]e agree with commenters that use of a plan’s own premium may cause
unintended distortions in transfers.” 2014 Final Rule, 78 Fed. Reg. at 15,432 (A.R.000249).14
What distortions HHS is describing is unexplained.

The sentence’s contexts suggest that

13

In subsequent final rules, HHS does not elaborate further on its budget-neutrality
rationale. See 79 Fed. Reg. at 13753-54 (A.R.004542-43); 80 Fed. Reg. at 10,771 (A.R.005703);
81 Fed. Reg. at 12,230 (A.R.007774); 81 Fed. Reg. at 94099-100 (A.R.009636-37). Those final
rules also do not provide additional rationales for using the statewide average premium. The
Court’s holding, thus, applies to the 2014, 2015, 2016, 2017, and 2018 rules.
14

HHS quotes another portion of the record in this argument, but, at that record portion,
the agency does not discuss why it is using the statewide average premium over an issuer’s own
premium. See Defendant Mem. at 21 (citing 77 Fed. Reg. at 73,140 (A.R.000135)). Instead that
record portion defines what a state average premium is and discusses some assumptions that the
agency made when deriving the payment transfer formula. See 77 Fed. Reg. at 73,140
(A.R.000135)(“The State average premium is the average premium requirement for providing
insurance to the applicable market population. . . . Finally, the derivation of the payment transfer
also assumed that plans price to cost, that is, that competition among plans for enrollees drives
plans’ premiums to their premium requirements.”). Neither quotation, in context, supports HHS’
contention that it reasons in the record that it selects the statewide average premium over an
issuer’s own premium, because the statewide average premium better calculates actuarial risk.
- 69 -

distortions refer to an imbalance in risk adjustment charge transfers, which implicates budget
neutrality and not that statewide average premiums better capture actuarial risk than an issuer’s
own premium. Alternatively, it could be a reference to HHS’ proposed rule’s prognostication
that “balancing adjustments could introduce differences in premiums across plans that were not
consistent with features of the plan.” 77 Fed. Reg. at 73,139. Again, HHS offers, in its briefing,
sound policy reasons for using the statewide average premium instead of the issuer’s own
premium, see, e.g., Defendant Mem. at 21 (“The state average also reduces the effect of
inaccurate or outlying pricing decisions by individual plans that could result in the methodology
under- or over-compensating for actuarial risk.”); the agency does not, however articulate those
reasons in the record.
Second, HHS argues that it selects the state average premium over an issuer’s own
premium “to reduce incentives for plans to avoid high risk enrollees.” Defendant Mem. at 21
(citing 78 Fed. Reg. at 15,411 (A.R.000228); White Paper at 36, 50 (A.R.000682, 000696)). The
record portions cited, however, do not articulate that policy reason. It is true that the Federal
Register cite notes that one of the ACA’s policy goal is to “reduce the incentives for issuers to
avoid higher-risk enrollees,” but it does not explain how a statewide average premium would
better effect that goal than an issuer’s own premium. 78 Fed. Reg. at 15,411 (A.R.000228). The
White Paper cited also does not explain why HHS selects the statewide average premium; rather,
it assumes that budget neutrality is a given and analyzes what effect that neutrality has on issuers
if HHS decided to use an issuer’s own premium. See White Paper at 36, 50 (A.R.000682,
A.R.000696). Because the record does not include HHS’ proffered rationale, the Court cannot
afford it deference. See Bowman Transp., Inc. v. Arkansas-Best Freight System, Inc., 419 U.S.
at 285-86. As the Court concludes that HHS has “failed to provide a reasoned explanation for its

- 70 -

action,” it sets aside and vacates the agency action as to the statewide average premium rules and
“remand[s] the case to the agency for further proceedings.” Olenhouse, 42 F.3d at 1575. See 5
U.S.C. § 706(2)(A)(“The court shall . . . hold unlawful and set aside agency action, findings and
conclusions found arbitrary [and] capricious.”).
Although the Court concludes that HHS’ 2014-2018 rules vis-à-vis use of the statewide
average premium are arbitrary and capricious, it notes that one of Health Connection’s
arguments against the 2018 rule is meritless. Health Connections argues that the 2018 proposed
rule, which still uses the statewide average premium, but reduces the amount by fourteen percent
to account for administrative costs, is also arbitrary and capricious, because the 2018 rule does
not adequately account for competition and innovation, and ignores the statutory command,
which, according to Health Connections, instructs HHS to charge a risk adjustment based only
on actuarial risk. See Plaintiff Mem. at 31-32. Health Connections’ arguments fail on those
grounds, because the statute does not require HHS to consider only actuarial risk when
calculating the risk adjustment charge. See 42 U.S.C. § 18063. Moreover, in its 2018 proposed
rule, HHS considers whether “inclusion of administrative costs in the Statewide average
premium harms efficient plans” and concludes that as “we noted in the 2017 Payment Notice and
White Paper . . . low cost plans do not necessarily indicate efficient plans. Should a plan be low
cost with low claims costs, it could be an indication of mispricing, as the issuer should be pricing
for average risk.” 81 Fed. Reg. at 94,099 (dated December 22, 2016)(A.R.009636). Based on
this record, the Court concludes that HHS considered Health Connections’ efficiency and
innovation concerns for the 2018 proposed rule, and, accordingly, HHS did not act arbitrarily or
capriciously on those grounds. See Motor Vehicle Mfrs. Ass’n of U.S., Inc., v. State Farm Mut.
Auto. Ins. Co., 463 U.S. 29, 43 (1983).

- 71 -

III.

HHS’ APPROACH TO PREDICTING COSTS FOR HCC AND NON-HCC
ELIGIBLE ENROLLEES IS NOT ARBITRARY AND CAPRICIOUS.
Health Connections argues that HHS’ risk adjustment formula under-predicts costs for

enrollees without HCCs, because the formula does not consider the costs that a non-HCC
enrollee incurs when using preventive care services or suffering unexpected health issues. See
Plaintiff Mem. at 41. Consequently, Health Connections contends that the flawed formula
“penalizes the enrollment of younger and healthier members needed to balance the risk pool and
avoid a ‘death spiral’ of rising medical costs.” Plaintiff Mem. at 42 (quoting King v. Burwell,
135 S. Ct. at 2486). HHS’ approach is not arbitrary and capricious, because, as the record
shows, HHS addressed the issue by considering relevant factors and provided reasoned bases for
its decisions.
To begin, when making its rules for 2014-2016, the information before HHS did not
clearly demonstrate that its risk adjustment formula was going to have problems predicting cost
for non-HCC and HCC enrollees. On the one hand, a Blue Cross and Blue Shield comment
warned that overlooking non-HCC enrollees’ costs could make HCC enrollees more profitable
for insurers, see BCBSA Detailed Comments on the Interim Final Rule with Comment: “Patient
Protection and Affordable Care Act; Amendments to the HHS Notice of Benefit and Payment
Parameters for 2014” [CMS-9964-IFC] at 5 (undated)(A.R.004330), but on the other hand, a
study determined that risk adjusters generally underestimate the costs associated with higher cost
individuals, see Ross Winkelman, Society of Actuaries, A Comparative Analysis of ClaimsBased Tools for Health Risk Assessment at 24 (dated April 20, 2007)(A.R.001261); Blue Cross
and Blue Shield Comments on the Proposed Rule: “Patient Protection and Affordable Care Act;
HHS Notice of Benefit and Payment Parameters for 2014” [CMS-9964-P] at 52 (dated
December 28, 2012)(A.R.003098). Given the mixed signals HHS received, it was not a clear

- 72 -

error of judgment to finalize its proposed risk adjustment formula.
As for making its 2017 and 2018 rulemaking, HHS responded to criticisms that its risk
adjustment formula underestimates costs for healthy enrollees and provided reasoned bases for
its decisions. For instance, HHS’ 2017 rules incorporated preventative services costs into its risk
adjustment formula. See Defendant Reply at 12; 81 Fed. Reg. at 12,218 (dated March 8, 2016)
(A.R.007762); Centers for Medicare & Medicaid Services Center for Consumer Information &
Insurance Oversight, HHS-Operated Risk Adjustment Methodology Meeting Discussion Paper at
33 (dated March 31, 2016)(A.R.009757). HHS explained that it
attempted to address the range between enrollees without HHCs and those with
HCCs by finalizing the incorporation of preventive services into our simulation of
plan liability. While overall this is not a very large effect, it does have a
noticeable effect on certain demographic subgroups, resulting in more accurate
payments for enrollees without HCCs.
81 Fed. Reg. at 12,218 (dated March 8, 2016)(A.R.007762). HHS also considered modifying its
risk adjustment model to more accurately predict costs for healthier enrollees. When proposing
its 2018 rules, HHS stated that it was
evaluating an approach in which we would directly adjust plan liability risk scores
outside of the model for [certain] subpopulations. For example, we could
potentially make an adjustment to the plan liability risk scores calculated through
the HHS risk adjustment models that would adjust for such an underprediction or
overprediction in actuarial risk by directly increasing low plan liability risk scores
and directly reducing high plan liability risk scores in order to better match the
relative risks of these subpopulations.
81 Fed. Reg. at 61,473 (dated September 6, 2016). HHS did not immediately make such
adjustments, however, because it believed that “there is a risk that such modifications could
unintentionally worsen model performance along other dimensions on which the model currently
performs well.” 81 Fed. Reg. at 61,473. Accordingly, HHS decided to “continu[e] to evaluate
the effect of these types of modifications on all aspects of the model’s performance before

- 73 -

choosing to implement such an approach, and would not implement these types of modifications
if we determined that doing so would have material unintended consequences for the model's
performance along other dimensions.” 81 Fed. Reg. at 61,473.
HHS reported, in its final rule for 2018, that it believed that modifications to its risk
adjustment model could improve its accuracy for low-cost enrollees, but added that it was still
considering the tradeoffs associated with such modifications. See 81 Fed. Reg. at 94,083. It also
provided a detailed summary of the myriad comments it received regarding adjustments to its
risk adjustment model. See 81 Fed. Reg. at 94,083. Given the disparate recommendations
contained in those comments, it was neither arbitrary nor capricious for HHS to exercise caution
instead of taking precipitous action.
Health Connections argues, nonetheless, that HHS acted arbitrarily and capriciously by
not responding to the Foster Memorandum specifically. See Plaintiff Mem. at 36 (“Incredibly,
HHS did not respond to Mr. Foster’s white paper at all, much less offer any reasoning or data to
explain why it was not adopting his detailed proposal.”). Health Connections suggests that it is
arbitrary to capricious to “ignore[] critical comments to proposed rules,” Plaintiff Mem. at 36,
but Health Connections’ reference to “critical comments” muddies the distinction between
comments as to specific proposals and comments as to general issues raised before an agency.
For the proposition that an agency ignoring “critical comments to proposed rules” is arbitrary
and capricious, Health Connections cites Allied Local & Reg’l Mfrs. Caucus v. U.S. E.P.A., 215
F.3d 61, 79-80 (D.C. Cir. 2000), but that case speaks to the implications of an agency ignoring
“significant points raised during the public comment period,” 215 F.3d at 80. Cf. NRDC v. EPA,
859 F.2d at 188 (“The fundamental purpose of the response requirement is, of course, to show
that the agency has indeed considered all significant points articulated by the public.”). The

- 74 -

question for the Court is whether HHS “entirely failed to consider an important aspect of the
problem,” Colo. Envtl. Coal. V. Dombeck, 185 F.3d at 1167, and/or whether HHS failed to
“articulate a reasoned basis for its conclusions,” Olenhouse, 42 F.3d at 1580. Regarding its
decision to not adopt proposals to address prediction bias by directly adjusting plan liability risk
scores -- which appears, to the Court, to be an oblique reference to the Foster Memorandum -HHS provides a reasoned basis for its decision, explaining:
We note that while we believe modifications of this type could improve the
model’s performance along this specific dimension, there is a risk that such
modifications could unintentionally worsen model performance along other
dimensions on which the model currently performs well. For this reason, we are
continuing to evaluate the effect of these types of modifications on all aspects of
the model’s performance before choosing to implement such an approach, and
would not implement these types of modifications if we determined that doing so
would have material unintended consequences for the model’s performance along
other dimensions.
81 Fed. Reg. at 61,473. See Foster Memorandum at 2 (“The adjusted risk scores can be
calculated using a simple formula based on a plan’s unadjusted risk score from the HHS-HCC
model and its actuarial value.”). Accordingly, HHS did not act arbitrarily and capriciously here.
IV.

HHS’ DECISIONS REGARDING PARTIAL YEAR ENROLLEES AND THE USE
OF PRESCRIPTION DRUG DATA ARE NOT ARBITRARY AND CAPRICIOUS.
The Court concludes that HHS’ decisions regarding partial year enrollees and the use of

prescription drug data in its risk adjustment model are not arbitrary and capricious. HHS’ risk
adjustment model calculates a risk score for every enrollee, which quantifies an enrollee’s
relative healthiness, and, therefore, the estimated costs of providing healthcare to that enrollee.
See Plaintiff Mem. at 40. HHS begins by considering an enrollee’s age and sex, but if HHS
learns that an enrollee qualifies for an HCC -- such as diabetes or HIV/AIDS -- the enrollee’s
risk score increases. See Plaintiff Mem. at 40-41. Health Connections argues that HHS does not
accurately identify enrollees who have HCCs because of how HHS accounts for partial year

- 75 -

enrollees and because HHS only uses diagnoses -- and not prescription drug data -- when
determining whether an enrollee has an HCC. See Plaintiff Mem. at 45. It follows, according to
Health Connections, that enrollees’ risk scores are inaccurate. See Plaintiff Mem. at 45. A
partial-year enrollee is someone who is enrolled in a health insurance plan for part of a year. See
Plaintiff Mem. at 45 (citing Hickey Declaration at 22 (NMHC000884)). If an enrollee has an
HCC, but is not diagnosed while enrolled in a particular plan, then that plan will not report the
HCC. See Plaintiff Mem. at 45 (citing Hickey Declaration at 22-23 (NMHC000884-85)). Using
prescription drug data would capture an enrollees HCC status even if that enrollee has not been
diagnosed with an HCC. For example, a person may have been diagnosed with diabetes before
enrolling in their current health insurance plan. That health plan, therefore, might not have a
record of their diabetes diagnosis. The health plan would, however, have records regarding this
hypothetical person’s insulin prescriptions -- because those prescriptions are filled on a regular
basis -- which would indirectly indicate a diabetes diagnosis. See Plaintiff Mem. at 46 (citing
Hickey Declaration at 25 (NMHC000887)). In short, Health Connections contends that HHS’
decisions regarding how to account for partial year enrollees and its decision not to use
prescription drug data to identify enrollees with HCCs are arbitrary and capricious. See Plaintiff
Mem. at 49. The Court concludes otherwise.
1.

Partial Year Enrollees.

HHS’ decision regarding how to account for partial year enrollees is not arbitrary and
capricious. HHS rationally considered and addressed concerns regarding partial year enrollment.
See 78 Fed. Reg. at 15,421 (dated March 11, 2013)(A.R.000238).

HHS notes that “[w]e

received several comments that the HHS risk adjustment models do not appropriately account for
short-term enrollment. One commenter suggested that risk scores for individuals that were

- 76 -

enrolled for only part of a year would be inaccurate.” 78 Fed. Reg. at 15,421 (dated March 11,
2013)(A.R.000238). HHS gave two responses to this comment. It first stated that “[o]ur models
were calibrated to account for short-term enrollment in several ways. First, enrollee diagnoses
were included from the time of enrollment.”

78 Fed. Reg. at 15,421 (dated March 11,

2013)(A.R.000238). This inclusion means that, “if an enrollee joined a plan in April but did not
receive her diagnosis until July, she was nevertheless treated as having the condition for the
entire period of enrollment,” meaning starting in April. Defendant Mem. at 37. HHS next stated
that,
in the statistical estimation strategy for the HHS HCCs, average monthly
expenditures were defined as the enrollee’s expenditures for the enrollment period
divided by the number of enrollment months, annualized expenditures (plan
liability) were defined as average monthly expenditures multiplied by 12, and
regressions were weighted by months of enrollment divided by 12. We believe that
this statistical strategy, alongside the minimum enrollment requirement, ensures
that monthly expenditures are correctly estimated for all individuals.
78 Fed. Reg. at 15,421 (dated March 11, 2013)(A.R.000238). Health Connections argues that
HHS’ first response, “while helpful, does not address the core problem of individuals with very
short enrollment periods who never see a doctor while enrolled.” Plaintiff Mem. at 48. Health
Connections argues that HHS’ second response “exacerbates the problem, because many
enrollees -- such as woman giving birth -- have their expenses concentrated in a small time
period, and thus averaging such expenses over twelve months significantly underestimates the
costs of partial year enrollees.”

Plaintiff Mem. at 48 (citing Hickey Declaration at 23

(NMHC000885)).
Notwithstanding Health Connections’ criticisms, by including an enrollee diagnosis from
the time of enrollment and averaging monthly expenditures, HHS addressed important aspects of
the partial-year enrollee problem, namely how to estimate a partial year enrollee’s healthcare

- 77 -

costs. See Colo. Envtl. Coal. v. Dombeck, 185 F.3d at 1167. Further, HHS’ statement that “[w]e
believe that this statistical strategy, alongside the minimum enrollment requirement, ensures that
monthly expenditures are correctly estimated for all individuals” is not “so implausible that it
could not be ascribed to a difference in view or the product of agency expertise.” Colo. Envtl.
Coal. v. Dombeck, 185 F.3d at 1167 (citations omitted). HHS’ statistical estimation strategy that
attempts to capture an enrollee’s healthcare expenses is the product of agency expertise. Colo.
Envtl. Coal. v. Dombeck, 185 F.3d at 1167.
Health Connections next argues that HHS later “received a number of comments
addressing the problem and offering solutions,” but that “it provided no analysis of the issues
raised.”

Plaintiff Mem. at 49 (citing 81 Fed. Reg. at 12,220 (dated March 8,

2016)(A.R.007764)). Instead, HHS responded that “[w]e appreciate commenters’ substantive
feedback on accounting for partial year enrollment in future recalibrations and will continue to
analyze this issue and include our findings in the White Paper for discussion at the March 31,
2016 risk adjustment conference.” 81 Fed. Reg. at 12,220 (dated March 8, 2016)(A.R.007764).
That HHS essentially stated that it would address new comments at a meeting does not render
HHS’ behavior arbitrary and capricious. Indeed, “[a]gencies are not required to consider every
alternative proposed nor respond to every comment made.” Ron Peterson Firearms, LLC v.
Jones, 760 F.3d 1147, 1163 (10th Cir. 2014)(internal quotations omitted). More importantly,
“[f]ailure to respond is not grounds for APA invalidation unless the points raised in the
comments were sufficiently central that agency silence would demonstrate the rulemaking to be
arbitrary and capricious.” NRDC v. EPA, 859 F.2d at 188. There was, however, no agency
silence. After the meeting, “HHS explained that it would recalibrate the model by adding
enrollment duration factors.”

Plaintiff Mem. at 49 (citing 81 Fed. Reg. at 94,072 (dated

- 78 -

December 22, 2016)(A.R.009609)). Given that HHS held a public meeting and changed its
model to one more favorable to Health Connections, the Court cannot soundly hold that HHS
acted arbitrarily and capriciously.
2.

Using Prescription Drug Data.

HHS’ decision not to use prescription drug data in its risk adjustment model was not
arbitrary and capricious. HHS specifically considered the advantages and disadvantages of using
prescription drug data.

See RTI International Memorandum at 4 (dated December 15,

2011)(A.R.000838). HHS understood that “the main advantage to drug data are its completeness
and timely availability . . . nearly all prescription activity creates an electronic record that is
rapidly available to insurance plans.” RTI International Memorandum at 2 (A.R.000836). HHS
considered, however, that “[c]ounterbalancing the advantages of drug data in risk adjustment
models are several problems. Chief among these is the incentives that drug-based payments
systems would create.” RTI International Memorandum at 4 (A.R.000838). Specifically, “[t]he
most salient concern with tying risk adjustment payments to drug usage is the likely distortion of
provider decisions toward pharmaceutical therapies.”

RTI International Memorandum at 4

(A.R.000838). “This distortion would create real costs: not only the costs . . . of the drugs
themselves, but also the health outcomes that would be diminished by any deviation from
clinical best practices.” RTI International Memorandum at 4 (A.R.000838). These costs may
include “the risk of doctors steering patients toward drugs rather than behavioral therapies, but
also a bias towards certain types of drugs associated with high cost conditions.” Defendant
Mem. at 40 (citing RTI International Memorandum at 6-7 (A.R.000840-41)).
Further, HHS adequately performed its duty to “consider and respond to significant
comments received during the period for public comment.” Perez v. Mortgage Bankers Ass’n,

- 79 -

135 S. Ct. 1199, 1204 (2015). HHS’ proposed rule for 2014 stated:
At this time, we have elected not to include prescription drug use as a predictor in
each HHS risk adjustment model. While use of particular prescription drugs may
be useful for predicting expenditures, we believe that inclusion of prescription
drug information could create adverse incentives to modify discretionary
prescribing. We seek comments on possible approaches for future versions of the
model to include prescription drug information while avoiding adverse incentives.
77 Fed. Reg. at 73,128 (dated December 7, 2012)(A.R.000123). Health Connections contends
that HHS did not adequately respond to comments regarding this proposed rule, such as the
Pharmaceutical Research and Manufacturers of America Comment (dated December 20,
2012)(A.R.002765-70). See Plaintiff Mem. at 47. That comment states that “prescription
medicine utilization can improve the precision of risk scores for certain conditions . . . .
[R]esearch has shown that including prescription utilization in . . . risk adjustment models
improves the correlation between risk score and actual claim costs.” Pharmaceutical Research
and Manufacturers of America Comment at 4 (A.R.002768). Essentially, this language argues
that the use of prescription drug data would make HHS’ risk adjustment model more accurate.
See Pharmaceutical Research and Manufacturers of America Comment at 4 (A.R.002768).
HHS’ 2014 proposed rule had already balanced, however, the possible accuracy that prescription
drug data may provide against other concerns, explaining that, “[w]hile use of particular
prescription drugs may be useful for predicting expenditures, we believe that inclusion of
prescription drug information could create adverse incentives to modify discretionary
prescribing.” 77 Fed. Reg. at 73,128 (dated December 7, 2012)(A.R.000123). HHS’ subsequent
responses therefore did not require further detail because HHS’ reasoning for its decision was
already “clear to the public.” NRDC v. EPA, 859 F.2d at 189.
HHS specifically considered the advantages and disadvantages of using prescription drug
data. It may be that using prescription drug data would have been a superior public policy

- 80 -

choice; indeed, HHS ultimately changed its mind and decided to use at least some prescription
drug data in its risk adjustment model beginning in 2018. See 81 Fed. Reg. at 94,076 (dated
December 22, 2016)(A.R.009613). See also Encino Motorcars, LLC v. Navarro, 136 S. Ct. at
2125 (“Agencies are free to change their existing policies as long as they provide a reasoned
explanation for the change.”). A suboptimal decision, however, is not necessarily arbitrary and
capricious, and the Court will not second-guess HHS’ reasonable decision not to incorporate
prescription drug data into its risk adjustment model.
V.

HHS’ RISK ADJUSTMENT FORMULA
ELIMINATE BRONZE LEVEL PLANS.

DOES

NOT

EFFECTIVELY

Health Connections argues that HHS’ risk adjustment methodology violates the ACA,
because it is so burdensome that it effectively eliminates bronze level plans while the ACA took
pains to create those plans. See Plaintiff Mem. at 50-51. The ACA states:
Levels of coverage defined
The levels of coverage described in this subsection are as follows:
(A) Bronze level
A plan in the bronze level shall provide a level of coverage that is
designed to provide benefits that are actuarially equivalent to 60
percent of the full actuarial value of the benefits provided under
the plan.
42 U.S.C. § 18022(d)(1)(A). By its plain language, that provision states that a bronze level plan
shall have certain features, but it is silent regarding the profitability of those plans. It does not,
for example, state that the plan must always be profitable. With its actions, the Court is not
convinced that HHS has, as Health Connections insists, “[d]e [f]acto [b]anned Bronze Plans.”
Plaintiff Mem. at 50. For instance, Health Connections argues that many insurers have stopped
offering bronze plans, but it does not argue that bronze have vanished entirely. See Plaintiff
Mem. at 51. Meanwhile, HHS asserts that “New Mexico health insurers proposed more than 20
- 81 -

different Bronze plans for the 2018 benefit year, including several offered by [Health
Connections].” Defendants’ Reply at 23. The trend Health Connections identifies raises serious
concerns, but those concerns do not mean that HHS has violated the ACA.
Health Connections asks the Court to “remand to the agency . . . [for HHS] to grapple
with the question of how the agency can prevent the risk adjustment program from gutting
Congress’s intent to have viable bronze product offerings,” Plaintiff Mem. at 44, or to order HHS
to respond to Health Connections’ comment about how HHS was going to adjust the formula so
bronze plans are not unfairly penalized, see Tr. at 112:9-15 (Bassman). Such relief is neither
necessary nor appropriate. HHS has responded to comments that its risk adjustment formula
disadvantages bronze level plans. In 2016, HHS addressed proposals for improving its risk
adjustment formula’s predictive ability and noted that “[a] few commentators . . . suggested that
bronze plans are . . . specifically disadvantaged by the existing risk adjustment model.” 81 Fed.
Reg. at 94,083 (dated December 22, 2016)(A.R.009620). HHS responded that, although some
suggestions may improve its risk adjustment formula’s predictive ability, it was “still evaluating
the tradeoffs that would need to be made in model predictive power among subgroups of
enrollees,” and it will “continue to explore these modeling approaches.” 81 Fed. Reg. at 94,083
(A.R.009620). The Court sees no basis for ordering HHS to provide an additional response to
this specific aspect of the discussion.
IT IS ORDERED that: (i) the Plaintiff’s Motion for Summary Judgment, filed April 13,
2017 (Doc. 32), is granted in part and denied in part; and (ii) the Defendants’ Cross-Motion for
Summary Judgment, filed June 1, 2017 (Doc. 34), is granted in part and denied in part. The
Court sets aside and vacates the agency action as to using a statewide average premium for the
2014, 2015, 2016, 2017, and 2018 rules and remands the case to the agency for further

- 82 -

proceedings. The Court dismisses the Plaintiff New Mexico Health Connections’ remaining
claims with prejudice.

________________________________
UNITED STATES DISTRICT JUDGE

Counsel:
Nancy Ruth Long
Long Komer & Associates, P.A.
Santa Fe, New Mexico
-- and -Barak A. Bassman
Leah Greenberg Katz
Marc D. Machlin
Sara Richman
Pepper Hamilton, LLP
Philadelphia, Pennsylvania
Attorneys for the Plaintiff
Chad A. Readler
Acting Assistant Attorney General
Joel McElvain
Assistant Branch Director
Arjun Garg
Serena Orloff
James R. Powers
Trial Attorneys
United States Department of Justice
Civil Division, Federal Programs Branch
Washington, D.C.
Attorneys for the Defendants

- 83 -