In comments last week, Mayken expressed a fairly common sentiment on Health Savings Accounts (HSAs)
My husband’s job is offering a HSA and he’s actually considering it. Scares the crap out of me frankly. As a person with a few chronic issues including asthma and a ridiculous number of meds plus a child under 10 at home, I can’t see how it can be good for us.
HSAs and High Deductible Health Plans (HDHP) to which they are attached to are not inherently scary. They may not be the optimal plan for everyone but they are a logical choice for some groups of people.
An HSA is a savings account that has incredible tax advantages:
- Deposits are tax free
- Growth is tax free
- Withdrawals are tax free if used for qualified medical expenses
An HSA can only be opened if an individual is covered by a HDHP which is a plan that has a deductible of at least $1,350/$2,700 (individual/family) with a maximum out of pocket of no more than $6,750/$13,500 (individual/family). A HDHP covers nothing before the deductible is met EXCEPT for the key no-cost sharing preventative services (mammograms, flu shots etc). There are high deductible plans which are not qualified high deductible plans as those plans will cover some services pre-deductible. The actuarial value of a HDHP ranges from 60% (Bronze) to 86% (Platinum) . It is more a benefit design and tax designation than an actuarial value designation.
The HDHP/HSA theory of change is that old style plans with low or no deductibles encouraged people to be price insensitive and to be willing to say yes to anything as they bore no immediate cost. Shifting the plan design so that an insured individual or family is on the hook for a significant chunk of the initial because of the deductible would lead to better shopping for either lower priced care for the same services or the elimination of low value services entirely.
That is not the reality. Brot-Goldberg et al found that HDHP/HSA plans led to a significant reduction in spending. That reduction was solely through the indiscriminate reduction in services and not through better shopping or elimination of low value or wasteful care. Haviland et al have found that HDHPs don’t produce snap-back spending three years out. We’re bad shoppers of complex goods and services where being wrong has potentially infinite costs.
From here, the theory of change extends a bit. Using a lifecycle model where most people are mostly healthy/cheap when young (pregnancies excluded), the idea is that people would build up big balances in HSAs during their 20s, 30s and 40s and then use those balances to pay high deductibles in their 50s and 60s. This is problematic for people who either hit their out of pocket maxes in most/all years at which point the HSA mainly acts as a tax washer on money and the HDHP acts as a permanent income tax based on disease burden/bad luck. It is also problematic for people who can’t put significant net sums into an HSA because they don’t earn enough. On average, it probably works but there are a lot of people left behind.
However, that is theory and general population insight; let’s talk specifics.
HDHPs/HSAs can be a good deal for some people.
The calculation people need to make is whether or not the Premiums + Deductible (realized) -Tax Advantages (Deductible spend (realized)) is greater or less than a similar calculation for other plans. Companies will often offer HDHPs with lower employee premium shares than non-HDHPs so there are plenty of situations where an HDHP/HSA combination can make sense.
However, even if the math makes sense over the course of a year, there are some caveats.
- Credit/consumption shifting has to happen in the first year
- Initial services require cash payment
- First few months of the year are likely to be all cash if you modest chronic conditions (asthma, diabetes etc)
- You see full contracted price
- More record keeping
HDHP/HSA are not panaceas. They are also not anathema.
Depending on your circumstances, risk tolerance and your ability to float a fairly large payment out of regular cash flow or other assets, they can make sense.