Will Divestment from Employment‐Based Health Insurance Save Employers Money? The Case of State and Local Governments

• Author: Jeremy D. Goldhaber‐Fiebert,Monica S. Farid,David M. Studdert,Jay Bhattacharya
• Published date: 01 September 2015
• DOI: http://doi.org/10.1111/jels.12076
• Date: 01 September 2015

Will Divestment from Employment-Based

Health Insurance Save Employers Money?

The Case of State and Local Governments

Jeremy D. Goldhaber-Fiebert, David M. Studdert, Monica S. Farid,

and Jay Bhattacharya*

Reforms introduced by the Affordable Care and Patient Protection Act (ACA) build new

sources of coverage around employment-based health insurance. But what if firms find it

cheaper to have their employees obtain insurance from these sources, even after

accounting for penalties (for nonprovision of insurance) and employee bonuses (to ensure

the shift is cost neutral for employees)? State and local governments (SLGs) have strong

incentives to consider the economics of such “divestment”; many have particularly large

unfunded benefits liabilities. We investigated whether SLGs would save under two

scenarios: (1) shifting all employees and under-65 retirees to alternative sources of

coverage and (2) shifting only employees whose household incomes indicate they would be

eligible for federally-subsidized coverage and all under-65 retirees. Full divestment would

cost SLGs more than they currently pay, due primarily to penalty costs. Selective divestment

could save SLGs nearly $129 billion over 10 years at the expense of the federal government.

I. Introduction

Sweeping national policy reforms often reset the division of fiscal responsibilities

between the public and private sectors, and between different levels of government

(Rodden 2006). They may also create strong incentives for cost shifting, particularly in

*Address correspondence to Jeremy D. Goldhaber-Fiebert, Assistant Professor of Medicine and of Health

Research and Policy (by courtesy), Stanford Health Policy, Centers for Health Policy and Primary Care and Out-

comes Research, Stanford University, 117 Encina Commons, Stanford, CA 94305-6019; email: jeremygf@stanford.

edu. Studdert is Professor of Medicine and of Law, Stanford University; Farid is Data Analyst/Programmer, Stan-

ford University; Bhattacharya is Professor of Medicine, of Economics (by courtesy), of Health Research and Policy

(by courtesy), and Senior Fellow of the Stanford Institute for Economic Policy Research and of the Freeman Spo-

gli Institute for International Studies (by courtesy), Stanford University.

The authors gratefully acknowledge Daniel R. Austin, Anna Luan, and Louise L. Wang, co-authors along with

Dr. Bhattacharya on an earlier paper in Health Affairs that examined the effect of shifts to employer insurance

premiums. These authors generously shared Stata code that formed the basis for the ACA subsidy and cost-

sharing calculations used in our work. The authors also wish to thank the Stanford Health Policy faculty and

trainees for insightful comments and questions during the development of this work. Dr. Goldhaber-Fiebert was

supported in part by the National Institutes of Health Nation Institute on Aging (K01-AG037593; PI: Goldhaber-

Fiebert). Dr. Bhattacharya also thanks the National Institute on Aging for support (P01-AG05842 and R37-

AG036791).

343

Journal of Empirical Legal Studies

Volume 12, Issue 3, 343–394, September 2015

immediate postreform periods when “loopholes” abound in evolving regulations

(Obinger & Castles 2005). The Patient Protection and Affordable Care Act (ACA) is no

exception. The federal government’s role in health-care financing expands substantially

as it takes on subsidies and cost sharing for lower-income households, as well as the

lion’s share of the costs of Medicaid expansions.

The enhanced role for federal government financing has the potential to chill

employers’ willingness to offer health insurance (Cutler & Gruber 1996). With coverage

for most workers and their families available through the health insurance exchanges or

expanded Medicaid programs, it may be appealing for firms to curtail their own offer-

ings. For this to be a cost-saving move, however, any savings from having employees

obtain insurance elsewhere would need to exceed penalties levied under the ACA.

1

Such health insurance “divestment” by employers is not what the ACA’s architects

intended, but the potential for firms to have financial incentives to pursue it cannot be

ignored.

One large group of employers with pressing reasons to consider health insurance

divestment is state and local government (SLG). SLGs were particularly hard hit by the

most recent financial crisis. Between 2009 and 2012, budget shortfalls among state gov-

ernments exceeded $540 billion (Oliff et al. 2012), prompting deep cuts to spending

and services. Although the fiscal position of SLG has generally improved since 2012,

many will take years to return to a stable financial position (Oliff et al. 2012).

Health insurance benefits constitute a substantial component of SLG budgets. In

2013, most SLGs offered health insurance to their employees, and many extended bene-

fits to retirees under age 65, but continuing to do so will be challenging (Kaiser Family

Foundation and Health Research and Education Trust 2013). A 2010 Pew Center report

estimated that states’ obligations to public-sector retiree health-care benefits exceeded

states’ assets by $627 billion (Pew Center on the United States 2012; GAO 2009; Govern-

mental Accounting Standards Board 2004). A more recent Pew Center report co-

commissioned with the MacArthur Foundation finds that SLGs offer more generous

insurance than large private employers, cover a greater proportion of their employees’

health-care costs, and are more likely to offer first-dollar coverage (Pew Charitable

Trusts 2014). Hence, the weight of SLG health-care obligations in their current form is

particularly heavy.

Would SLGs alleviate their budgetary pressures by shifting current and retired

employees into federally-subsidized health insurance plans? If so, how much money

might they save? We addressed these questions by combining and analyzing data from

three national surveys. We began by estimating the costs to SLGs of continuing to offer

1

In addition to the incentives of capturing subsidies and cost sharing, an important example of how the ACA

may raise the cost of maintaining employer-sponsored insurance plans relative to cost prior to the ACA is the so-

called Cadillac tax that taxes overly generous plans. Press coverage has highlighted the risk of SLG insurance

plans in facing the Cadillac tax, and it is likely given SLG health-care plan generosity that it will be felt there first.

Yet, because the threshold at which the Cadillac tax kicks in is indexed to the overall CPI and medical costs have

historically grown much more quickly than the overall CPI, it is quite possible that many employers---even non-

SLGs---will eventually face the Cadillac tax themselves and hence greater incentives to shift away from employer-

sponsored coverage or at least to reduce the generosity of the plans they offer.

344 Goldhaber-Fiebert et al.

coverage in the usual way. Next, we estimated whether SLGs would save by dropping

employment-based insurance entirely and supporting employees and retirees to obtain

coverage elsewhere. We then estimated potential savings from selectively shifting defined

subgroups of beneficiaries to alternative coverage sources. Finally, we considered possi-

ble legal barriers to these moves by SLGs, particularly the selective shifting of current

employees onto the ACA exchanges.

Our findings should provide useful information to SLGs that are considering their

options in the wake of the ACA. Since savings to SLGs are achieved largely by shifting

costs to the federal government, our estimates are also relevant to ongoing debates

about projected costs of ACA-related reforms to the federal government.

A. The Divestment Calculus

The calculus of divestment is difficult for any employer and may be particularly so for

SLGs. It involves evaluating a complex interplay between financial and nonfinancial con-

siderations, including labor relations and supply issues, political climates, and the poten-

tial for policies to be changed through federal regulatory, legislative, or legal processes.

The ACA does not establish special rules for SLGs; essentially, the law treats them

like any other employer.

2

Thus, under the ACA’s “play-or-pay” regime, if an SLG with at

least 50 full-time equivalent workers does not offer at least one coverage option to every

full-time employee, and an employee obtains subsidized coverage through an individual

exchange, the employer must pay a $2,000 penalty for each full-time employee above

the first 30. Alternatively, if an SLG with 50 full-time equivalent employees offers cover-

age but employees choose instead to purchase insurance on the exchange, the SLG

faces a penalty if the purchase attracts a subsidy. In this case, the SLG faces the lesser of

two penalties: $3,000 per employee who purchases subsidized coverage or $2,000 for all

full-time employees above the first 30.

Nor does the ACA stipulate special rules or exceptions for SLG employees. They

may purchase health insurance offered on the exchange, or enroll in Medicaid if they

meet eligibility criteria. Income thresholds for Medicaid eligibility vary by state, but the

ACA sets a floor: states implementing the expansions must accept households with

incomes below 138 percent of the federal poverty level (FPL), which in 2013 corre-

sponded to $15,856 for an individual and $26,951 for a family of three. Households

with incomes between 138 percent and 400 percent FPL are eligible for subsidies and

cost sharing for exchange-purchased plans, and these are calculated on a sliding scale.

3

For SLGs and other employers, shifting employees to Medicaid and exchange-

purchased plans may be fiscally attractive. However, the financial calculus is not straight-

forward. It depends on several factors—principally, how many employees are Medicaid

eligible, how many are eligible for subsidies and cost sharing (and at what level), and

2

Rules described in this section come from various provisions of the Patient Protection and Affordable Care Act.

3

The ACA actually specifies 133 percent of the federal poverty level as the Medicaid eligibility threshold, but the

first 5 percentage points of income are disregarded, effectively making it 138 percent.

345Divestment from Employment-Based Health Insurance