Employer Decisions to Self-Insure
Does State Regulation Matter?
About This Event

Why do some employers decide to self-insure for their health insurance plans while others do not? Are they influenced by the fact that state laws prohibit managed care plans from restricting the employee's choice of provider? What effect would congressional or court-ordered changes to the federal law ERISA have on the future of employer health insurance?

In a new study sponsored by the Health Care Financing and Organization Program of the Robert Wood Johnson Foundation, Gail Jensen and Michael Morrisey use data from the 1990s on employer-sponsored health insurance plans and changes in state health insurance laws to study the effects of state regulation and other market forces on employers' decisions to self-insure. When an employer decides to self-insure for its group health insurance, it foregoes participating in an insurance pool organized through a health insurance company and instead accepts the responsibility of managing its own risks. Under ERISA, self-insured plans are exempt from most state insurance requirements and state premium taxes. The dramatic increase in self-insurance in the early 1990s coincided with the growth of a variety of managed care plans, which varied widely throughout the states

Has the proliferation of state laws targeted at managed care plans motivated employers to self-insure? At this event, Gail Jensen will present findings about state managed care regulation and the expansion of self-insurance among employers. Participants will also discuss the policy implications of these findings.

8:45 a.m.


9:15 Introduction: Anne K. Gauthier, AcademyHealth
Presentation: Gail Jensen, Wayne State University
Discussion: Roland McDevitt, Watson Wyatt
David Hyman, University of Maryland and University of Illinois
Moderator: Robert B. Helms, AEI


Event Summary

June 2004

Employer Decisions to Self-Insure

Why do some employers decide to self-insure for their health insurance plans while others do not? Are they influenced by the fact that state laws prohibit managed care plans from restricting the employee's choice of provider? What effect would congressional or court-ordered changes to the federal law ERISA have on the future of employer health insurance?  At a June 2 AEI conference, Gail Jensen presented the findings of a study sponsored by the Health Care Financing and Organization Program of the Robert Wood Johnson Foundation. Data from the 1990s on employer-sponsored health insurance plans and changes in state health insurance laws were used to study the effects of state regulation and other market forces on employers' decisions to self-insure.  

Gail Jensen
Wayne State University

With colleagues Michael Morrisey and Jon Gabel, I have investigated firms' decisions to self-insure in the context of state and federal regulation. Our findings will be published in a forthcoming study.

Self-insurance is vital to understand because it affects the ability of state laws to protect and benefit subscribers.  States generally oversee commercial health plans, setting financial reserve requirements and premium taxes to ensure the solvency of plans.  States also tax plans to finance risk pools for uninsurable populations, mandate essential services for plans, and regulate underwriting practices.  However, under the Employee Retirement Income Security Act of 1974 (ERISA), states have no oversight of self-insured plans, and there is limited federal oversight.

The largest impact of state regulation is seen in mandated benefits.  While they can favorably enhance benefits, compulsory coverage of services that purchasers would not otherwise insure increases the cost of the plan and, subsequently, premiums.  There are other less-obvious consequences of state mandated benefits, like increases in self-insurance, reduced provision of health benefits by employers, decreases in workers' wages, or cutbacks in non-mandated areas. 

Previous studies have compared the efficiency of self-insured plans and purchased plans.  Some research suggests that even for comparable coverage, self-insured plans are less efficient.  Still, firms continue to self-insure.  ERISA has resulted in two separate but unequal regulatory systems.  Purchased plans are subject to a large number of state regulations and mandated benefits.  However, under ERISA, self-insured plans have only limited state and federal oversight.  The ability of self-insured firms to avoid high-priced, state-mandated benefits offsets the added loading costs of self-insurance and maintains marketplace interest in this option.

The 1990s were a period of change for health insurance and its regulation. Managed care plans, state mandated managed care regulations, and self-insurance all increased throughout the decade.  In the early 1990s, managed care plans grew rapidly.  In 1993, conventional fee-for-service plans held at least half of the employer market, but by the end of the decade, their market share had fallen to approximately 8 percent. 

Along with the change in market composition came an increase in state regulations targeted at managed care plans.  While few states had managed care regulations in 1993, the number dramatically increased by the end of the decade.  The new regulations governed access to specialists, payment of providers, external review of care denials, formal grievance procedures, and point-of-service (POS) options.  During this time, there was also an increase in the number of states enacting Any Willing Provider (AWP) and Freedom Of Choice (FOC) laws.

The number of firms choosing to self-insure increased throughout the 1990s.  Some believed that the rise of managed care would limit the number of firms choosing to self-insure.  However, the number of firms self-insuring continued to grow, merely incorporating and self-insuring the new managed care plans.  Between 1993 and 1999, the number of enrollees in a self-insured HMO grew from 5 percent to 20 percent.  The number of enrollees in a self-insured Preferred Provider Organization (PPO) went from 50 percent in 1991 to 67 percent in 1999.  The number of enrollees in a self-insured POS plan went from 29 percent in 1991 to 47 percent in 1999.  The net result of these increases is a continued preference for self-insurance.  By the end of the decade, roughly half of all persons with employer coverage participated in a self-insured plan.  

A firm's decision to self-insure is part of its broader strategy to compensate its workers.  A firm compares the perceived full cost of compensation with purchased coverage to the perceived full cost of compensation under self-insurance, and opts to self-insure when that cost is lower.  The full cost of compensation must account for the risk that is borne by self-insured plans.  If self-insurance allows a firm to avoid unwanted benefits and premium taxes, it can customize its coverage to meet worker preferences.  On the other hand, the firm must then bear the uncertainty of claims and administrative costs. 

The data for this analysis includes 5,212 firms, each with two hundred or more employees, and a large sampling of various plan types (conventional plans, HMOs, PPOs, and POS plans).  The data spans from1993 to 1999. It also captures information on state insurance regulation, state mandated benefits, and taxation.

At the firm level, an econometric model was used to estimate a firm's decision to self-insure its health plan.  The probability of self-insurance is a function of firm characteristics, market determinants, regulatory variables, and state-level effects. 

At the plan level, a different model was used to estimate the decision to self-insure a particular type of plan.  For the first time, separate econometric equations have been estimated for conventional, HMO, PPO, and POS plans.             

The results show that at the firm level, the decision to self-insure was highly sensitive to the inclusion of characteristics of a state that have been fixed over time, such as demographics and political environment.  Without these state-level fixed effects, the significance of state insurance regulations was greatly exaggerated.  Controlling for these state-level factors, the impacts of state regulations become insignificant in the decision to self-insure.    

The data show that state insurance regulation had a minimal effect on the decision to self-insure among specific types of plans as well. If regulation does not impact decisions to self-insure, what does?  The major determinant of self-insurance is whether a firm has more than one physical location.  Multi-site firms are twice as likely to self-insure their health plans.  The decision to self-insure is also more likely for larger firms, firms offering multiple plans, and firms that experience smaller premium increases.  Also, firms in the mining, manufacturing, wholesale trade, and transportation industries are more likely to self-insure.  At the plan-level, industry effects are mainly seen for PPOs but not for other types of plans.  

There is also strong evidence of "carry-over" in the decision to self-insure newer health plans.  Firms that self-insure a conventional health plan tend to self-insure managed care plans as well. 

Self-insurance remains widespread, despite the growth of managed care plans.  Evidence suggests that state regulations had no effect on the decision to self-insure throughout the 1990s.  HIPAA and other federal mandates enacted in 1996 also had little effect on self-insurance.   

Roland McDevitt
Watson Wyatt

After working at Watson Wyatt, I approach the question of self-insurance from the employee benefits consulting perspective. 

The study asks, "Does state regulation matter?"  State regulation matters not because firms are trying to get around state mandates and offer less generous benefits, but because they are trying to avoid the complexity of fifty different state systems.

Firms often self-insure to simplify the benefits they offer.  A large employer (with five thousand or more employees) typically wants a core national plan with a uniform benefits package. 

A core national plan with standard benefits has many advantages: administration is simpler, Human Resources staff can better understand the offering packages, and employees do not have to know about many different plans.  Simplified benefits are also more efficient and create fewer communication problems. 

The decision to self-insure is not a function of generosity.  There is ample evidence that national employers with more than a thousand employees are not self-insuring to offer less generous packages.  They may, however, choose not to cover such things as herbal medicines and extended chiropractic care.  Self-insuring simply allows them to control the benefits they offer. 
Employers throughout the 1990s often felt they were not getting enough value with commercial plans.  For these plans, administration costs and retention fees could be as high as 20 percent of plan costs. Many firms believed they could administer their own benefits packages for less. Also, continued double-digit cost growth in the 1990s did not give firms confidence in commercial options.  Moreover, buying health coverage commercially does not reduce the risk borne by firms; it is passed back to the firm one way or another. 

Size and plan types also motivate a firm's decision to self-insure. Professor Jensen earlier determined that 60 percent of firms with more than two hundred employees self-insured, while only 20 percent of firms with fewer than two hundred employees did.  At the plan level, 60 percent of firms offering PPO plans self-insured; only 20 percent of firms offering HMOs self-insured.        

David A. Hyman
University of Maryland

Recasting the previous observations, one can derive three different perspectives on why firms decide to self-insure.   

The first is as a response to inefficient regulation.  If the regulation were effective, a firm would not need to self-insure-it would simply comply.  The diversity of terms across states could motivate firms to self-insure, or the terms themselves could influence that decision.  The data complicates this distinction because it is difficult to differentiate between multi-site, multi-state firms, and those with multiple locations within one state.  Gail Jensen's research suggests that, in fact, it is the diversity of terms across states that leads firms to self-fund their health plans. 

Another aspect of inefficient regulation is the disparity between state and federal regulatory costs.  This may have to do with enforcement of the regulations.  Firms may also self-insure but not necessarily to align their plans with more or less restrictive state or federal requirements.  While some firms have chosen self-funding in order to save money, others have far exceeded the requirements of even the most restrictive states.  As a result, the employment-based market has not become a drive to the bottom when it comes to coverage terms. 

"Path dependence" may also explain firms' decisions to self-insure.  Firms are reluctant to change their mode of operation and incur significant transaction costs in the move from self-insured to insured markets, and vice versa.  Gail Jensen's data showed that firms with self-insured indemnity plans tended to also self-insure their managed care plans.  Continuing on the same path of operation is easier and less costly.  Closely related to path dependence and an interesting aspect of self-insurance is its prevalence in mining, manufacturing, transportation, and wholesale trade.  This suggests industry-specific trends, which could be driven by unionization.  It is also possible that industry-specific trends are the result of simply "following the crowd." 

The decision on how to provide health insurance comes down to making it or buying it.  As a result, it is not surprising to note that increased size correlates to self-funding.  However, it is hard to determine whether this is because large firms have more predictable claims behaviors or because they have a greater ability to invest in Human Resources.  Even with a third-party administrator (TPA), there must be a minimum HR capacity-a requirement more adequately met by larger firms.  TPAs and limited coverage by commercial insurance plans make self-funding more appealing to smaller firms.

The decision to self-insure has several policy implications.  Along with the state and federal statutes, the interpretation of ERISA directly affects the decision to self-insure.  The Supreme Court decision two years ago on Kentucky's AWP statutes could largely dissolve AWP statutes across the board.  In addition, harmonization of state terms will dampen self-funding, regardless of efficiency. 

AEI intern Nikhil Pandey prepared this summary.

View complete summary.
Also Visit
AEIdeas Blog The American Magazine

What's new on AEI

AEI Election Watch 2014: What will happen and why it matters
image A nation divided by marriage
image Teaching reform
image Socialist party pushing $20 minimum wage defends $13-an-hour job listing
AEI Participants


Robert B.
AEI on Facebook