Report on HSAs and Health Reimbursement Arrangements

Assets, Account Balances, and Rollovers, 2006–2014

by Paul Fronstin, Ph.D., Employee Benefit Research Institute (EBRI), and Anne Elmlinger, Greenwald & Associates. Excerpted from the January 2015 EBRI Issue Brief and reprinted with permnission.

 

Employers first started offering account-based health plans in 2001, when a handful of employers began to offer health reimbursement arrangements (HRAs), employer-funded health plans that reimburse workers for qualified medical expenses. In 2004, employers were able to start offering health plans with health savings accounts (HSAs), tax-exempt trusts or custodial accounts used to pay for health care expenses.

The theory was that these accounts would give individuals more control over funds allocated for health care services and would cause them to become more engaged in their health care, thereby using health care services more effectively, especially once they became more educated about the actual price of health services. Furthermore, these accounts could be used as tax-advantaged vehicles to save for future health care expenses.

By 2014, 27 percent of employers with 10–499 workers and 48 percent of employers with 500 or more workers offered either an HRA- or HSA-eligible plan.1 As a result, these plans, collectively known as consumer-driven health plans (CDHPs), covered about 26 million people in 2014, representing about 15 percent of the privately insured market (Fronstin & Elmlinger, 2014). As the number of people with a CDHP grows, total assets associated with these plans can be expected to grow as well.

See the complete Brief, with charts and graphs, here.

CDHPs are no longer a new type of health plan. HSAs have been around now for a decade and HRAs since 2001. Hence, a growing percentage of the population has had either an HRA or HSA for a number of years. In 2008, 19 percent of the population with an HRA or HSA had held an account for three to four years, and 4 percent for five years or more (Figure 1). By 2014, 27 percent had held an account for three to four years, and 13 percent for five years or more. As the length of time individuals have these plans increases, average account balances should increase as well.

This Issue Brief presents findings from the 2014 EBRI/Greenwald & Associates Consumer Engagement in Health Care Survey (CEHCS). This study is based on an online survey of 3,887 privately insured adults ages 21−64. The survey was designed to provide nationally representative data regarding the growth of CDHPs and high-deductible health plans (HDHPs) and the impact of these plans and consumer engagement more generally on the behavior and attitudes of adults with private health insurance coverage.

The remainder of this report examines the findings from the 2014 CEHCS as they relate to the level of assets held in HSAs and HRAs, account balances, and rollover amounts. Findings from this survey are compared with findings from the 2005‒2007 EBRI/Commonwealth Fund Consumerism in Health Care Survey, and the 2008‒2013 CEHCS.

Enrollment in CDHPs

Individuals with a deductible of at least $1,250 for individual coverage or $2,500 for family coverage who also have an HRA, or were enrolled in an HSA-eligible health plan (regardless of whether the HSA was opened) were assigned to the CDHP group in the CEHCS. Everyone else with a deductible of at least $1,250 for individual coverage or $2,500 for family coverage was assigned to the HDHP group. Individuals with no deductible or a deductible below $1,250 for individual coverage and $2,500 for family coverage were assigned to the “traditional” coverage category.

More detail about the methodology is provided in Fronstin & Elmlinger (2014). As reported in Fronstin & Elmlinger (2014), this survey finds that 14.7 percent of the population was enrolled in a CDHP in 2014, compared to 16.6 percent in 2013 (Figure 2). The difference between the 2013 and 2014 estimates is not statistically significant. However, the lower 2014 enrollment estimate affects other trend data presented in this paper. As a result of the lower estimate, it appears that total assets held in HSAs and HRAs fell, and that the growth rate in assets was negative as well, when, in fact, this is simply a function of a lower estimate for HRA and HSA enrollment between 2013 and 2014 that is within the margin of error.

Assets

According to findings from the 2014 CEHCS, there was $22.1 billion in HSAs and HRAs in 2014, spread across 10.6 million accounts (Figure 3).4 In 2008, there were only 4.2 million accounts with $5.7 billion in assets.5 In 2014, growth rates were negative for both assets and the number of accounts (Figure 4). As noted earlier, the decline between 2013 and 2014 was likely the result of the lower incidence rate of CDHPs in 2014 that was not statistically significant. Note that in 2013, the number of accounts grew 3 percent while assets grew 38 percent. Part of the reason for the difference in the growth of assets compared with the growth of accounts was due to a change in the survey. In prior years, “$3,000 or more” was the highest account balance category that a respondent could choose. Starting with the 2013 survey, respondents could report account balances in the “$5,000–$9,999” and “$10,000 or more” ranges.

Account Balances

In 2014, average account balances reached $2,077 (Figure 5). In 2008, average account balances were $1,356, and as recently as 2012 they were $1,534. Between 2008 and 2014, the percentage of accounts with a balance of $3,000 or more increased from 18 percent to 27 percent. At the same time, the percentage with a zero balance fell slightly from 9 percent to 7 percent (Figure 6).

  • Length of Time With Account
    Not surprisingly, the length of time that an individual has held an account had a major impact on the amount of money in the account. In 2014, the average account balance was $796 among people who held an account for less than six months, while those with an account six months to less than a year had an average of $1,423 (Figure 7). In comparison, individuals who had held an account for one to two years had an average of $2,499. Those who had held an account for three to four years had an average of $3,225. And those who had held an account for at least five years had an average account balance of $3,092.
  • Employer and Individual Contributions
    Annual contribution amounts, whether they came from the employer (in the case of both HRAs and HSAs), or from individuals (as they apply to HSAs only) had a strong impact on overall account balances. Not surprisingly, the more money that was contributed to an account, the higher the average account balance. In 2014, for instance, accounts with an employer contribution had a higher average balance than those without an employer contribution. Accounts with an employer contribution had an average balance of $2,403, whereas those without an employer contribution had an average balance of $2,056 (Figure 8).

    In 2014, average account balances reached $2,077. In 2008, average account balances were $1,356, and as recently as 2012 they were $1,534

    When it came to the level of the employer contribution, individuals with an employer contribution of less than $1,000 had an average account balance of $2,183 in 2014, while those with an employer contribution of at least $1,000 had an average of $2,768 in their account (Figure 9). Similarly, individuals who contributed less than $1,000 had an average account balance of $1,548, while those who contributed at least $1,000 had an average balance of $3,151 (Figure 10). It will be important to track this trend over time. Currently, account balances are low and are therefore invested in relatively safe vehicles such as money market funds (currently, investments are usually restricted to a money market fund until the savings account reaches a minimum threshold, such as $2,000 or $3,000). As account balances grow, the potential to invest in more risky investment vehicles (such as mutual funds and stocks) will grow. The opportunities for capital appreciation will increase but so will the opportunities for capital losses, even among individuals with high levels of employer and individual contributions.

  • Rollovers
    Like contribution levels, annual rollover amounts had a large impact on average account balances. Individuals with less than a $1,000 rollover had an average account balance of $1,452 (Figure 11). In comparison, individuals with a rollover of at least $1,000 had an average account balance of $4,792.

Rollovers

Employers have a tremendous amount of flexibility in designing health plans that incorporate an HRA. Leftover funds at the end of each year can be carried over to the following year (at the employer’s discretion). Employers can, however, place restrictions on the amount that can be carried over. When it comes to HSAs, any money left in an account at the end of the year automatically rolls over and is available in the following year because there is no use-it-or-lose-it rule.6 Overall, the percentage of individuals with a rollover has increased. In 2008, 16 percent of individuals with an HRA or HSA did not roll over any money (Figure 12).

By 2014, 11 percent did not have a rollover. The number of accounts with a rollover increased annually until 2013, and the total level of assets being rolled over has increased in all years except for 2010, 2013, and 2014. In 2008, 2.5 million accounts rolled over $2.5 billion (Figure 13). By 2014, 7.1 million accounts rolled over $8.9 billion. The average rollover increased from $1,000 in 2008 to $1,244 in 2014 (Figure 14).

  • Length of Time With Account
    The length of time that an individual has held the account had an impact on rollover amounts. The analysis found that people holding an account for one to two years had an average rollover of $982 in 2014 (Figure 15). In comparison, those holding an account for three to four years had an average rollover of $1,421, and those with an account at least 5 years old had an average rollover of $1,428.
  • Employer and Individual Contributions—
    Accounts with an employer contribution had a higher amount rolled over than those without an employer contribution in 2014. Accounts with an employer contribution had an average rollover of $1,280 in 2014, whereas those without an employer contribution had an average rollover of $1,069 (Figure 16). Individuals with an employer contribution of less than $1,000 had an average rollover of $1,211 in 2014, while those with an employer contribution of at least $1,000 had an average rollover of $1,314 (Figure 17). In contrast, individuals who contributed less than $1,000 had an average rollover of $961, while those who contributed at least $1,000 had an average rollover of $1,444 (Figure 18).

Conclusion

Employers first started offering health reimbursement arrangements (HRAs) in 2001, and they were able to start offering health plans with health savings accounts (HSAs) in 2004. In 2014, 27 percent of employers with 10–499 workers and 48 percent of employers with 500 or more workers offered either an HRA- or HSA-eligible plan. As a result, these plans covered about 26 million people in 2014, representing about 15 percent of the privately insured market. According to findings from the 2014 EBRI/Greenwald & Associates Consumer Engagement in Health Care Survey (CEHCS), there was $22.1 billion in HSAs and HRAs in 2014, spread across 10.6 million accounts. Overall account balances averaged $2,077 in 2014.

Over 7 million accounts rolled over money between 2013 and 2014, with $8.9 billion rolled over. The average amount rolled over was $1,244 in 2014.

References

Fronstin, P., & Elmlinger, A. "Findings From the 2014 EBRI/Greenwald & Associates Consumer Engagement in Health Care Survey." EBRI Issue Brief no. 407 (Employee Benefit Research Institute, December 2014).

Taylor, H. (2003). "Does Internet Research ‘Work’? Comparing Online Survey Results With Telephone Surveys." International Journal of Market Research , 42, no. 1 (August 2003).

Endnotes 1 See http://www.mercer.com/content/mercer/global/all/en/newsroom/modest-health-benefit-cost-growth-continues-asconsumerism-kicks-into-high-gear.html

2 See http://www.i-say.com/

3 In theory, a random sample of 2,000 yields a statistical precision of plus or minus 2.2 percentage points (with 95 percent confidence) of what the results would be if the entire population ages 21–64 with private health insurance coverage was surveyed with complete accuracy. There are also other possible sources of error in all surveys that may be more serious than theoretical calculations of sampling error. These include refusals to be interviewed and other forms of nonresponse, the effects of question wording and question order, and screening. While attempts are made to minimize these factors, it is impossible to quantify the errors that may result from them.

4 The CEHCS undercounts the number of HSAs and total HSA assets because the survey only examines HSAs among people with an HSA-eligible health plan. Individuals who have an HSA but not an HSA-eligible health plan are not examined in the survey.

5 The term “assets” is used loosely as it relates to HRAs. An HRA is typically set up as a notional arrangement and exists only on paper. Employees may view the HRA as if money was actually being deposited into an account, and they may carry a debit card that can be used to pay for health care services at the point of service, but employers do not incur expenses associated with the arrangement until an employee incurs a claim.

6 Individuals are also able to roll over funds from one HSA into another HSA without subjecting the distribution to income and penalty taxes as long as the rollover does not exceed 60 days. Rollover contributions from Archer Medical Savings Accounts (MSAs) are also permitted.