« "Like Most People?" | Main | WWRD? »

November 30, 2006

Imaginary HSAs

I'll take the posting of the latest Health Wonk Review to belatedly respond to Bryan Caplan's snarky argument that HSA's will too bring down the costs of health care, because it's basic economics.  Much of his post relies on some analogy about traveling to China, but I tend to think chemotherapy is different from jaunts to Shanghai in obvious enough ways that it's not a terribly useful metaphor. 

What Bryan actually argues is that if everyone had to pay the first $10,000 out of pocket, they would get less care.  Problem is, Bryan doesn't mention or doesn't know that the legal limit for a personal HSA's deductible is $5,200.  Only when you're dealing with a family does it go up to $10,200.  Butwaitthere'smore!  Those are maximum deductibles.  The average deductibles are far lower: $2,011 for an individual, $4,008 for a family.  You're going to tell me this is where the savings are, given that the costs are among rare individuals spending in the five or six digits?  I don't buy it.  Those folks will bust through the cap on their first surgical consult, and this doesn't disincentivize their care at all.

Now, Bryan isn't a health care expert, he's an economist, and from that perspective, what he's actually arguing is correct: If you make people pay a lot of money for their health care, they'll use less of it.  The problem is, HSA's don't do much of that for the costliest patients -- they save money on low level treatments, disincentivizing an array of basic care services.  Here's an interesting result from the recent Kaiser survey:


Oddly, HSA users arer less satisfied with the cost of their coverage -- which are cheaper -- than those in more traditional health plans, 41% to 60%, which strikes me as a bad sign for HSAs.  Bit I digress.  Since Bryan and I agree that the costs in the health system are on the higher end and is using a $10,000 deductible as his base, he wasn't talking about actual HSAs (I was), but imaginary ones with higher deductibles -- like the plan his coblogger proposes, with deductibles in the $30,000s.  The problem is, such plans aren't adopted, even in their weak forms (we don't even use the maximum deductibles now), because they're considered, rightly or wrongly, a cruel and counterproductive approach to health care.  They will, however, save a lot of money.  I'm happy to argue that we shouldn't go down that path for a variety of reasons, but it's a different argument than whether HSAs, as currently composed, will significantly lower costs.

November 30, 2006 in Consumer-Directed Health Care | Permalink



One quick note - the $5,100 and $10,200 limits are out-of-pocket limits, not deductible limits.

So, theoretically, you could have a $1,000 deductible (legal minimum for HSA qualification), and, then a 1% co-insurance rate spread over the next $410,000 before you run into your $5,100 out of pocket limit.

So, you'd be "more price sensitive" over every $ value of health spending from $0-410,000. Of course, a 1% co-insurance rate is silly, and, the general part of your argument still holds, but, HSA defenders like to argue that these out of pocket limits can, with modest co-insurance rates, be spread over very large values, and, can make people price-sensitive over that whole range.

Of course, the larger the range of medical costs you decide to make price sensitive, the lower the co-insurance rate, and, hence the smaller this price sensitivity.


Posted by: joshb | Nov 30, 2006 4:45:49 PM

The price sensitivity in the Kaiser graph appears to reflect a choice between getting medical care and not getting medical care. I would think to reduce costs (the acutal cost of medical care, rather than the cost to the purchaser) the individual would have to be able to choose their caregiver (doctor, hospital) based on cost which you cannot do as the patient does not get to negotiate, it is the insurance company that sets the price. Possibly if there were a choice of entire health care systems that you could choose based on price and services when you were well, that would effect the cost. i.e, I'd like to sign up with VA care, or Canada care or Sweden care. But we take the plan our employer offers. The only choice we have as medical care consumers is get care or don't get care. And please note - many people without HSA's have high co-pays. My plan calls for $500.00/day co-pay up to $2000.00 per hospital stay. If I'm in situation where I must be hospitalized - I'm not saying no - it's too expensive, I'm saying I'll deal with it later when the pain stops.

Posted by: Cathy | Nov 30, 2006 5:24:55 PM

Just to emphasize one of your points -- the Kaiser graph shows that HSAs work. They disincentivize health spending. Unfortunately, few if any of us know which test we miss, or which prescription we don't fill, will have repercussions later.

HSAs would make some sense if most health care spending was frivolous, but I haven't read anybody arguing to that effect...

Posted by: tinman | Nov 30, 2006 6:08:15 PM

The correct way to think about HSAs is that your put off early-in-disease spending on health care (when through early diagnosis and treatment you might be spared later serious problems) thus 'saving your health' until your health is nearly or all gone. Then the insurer will come along and attempt to restore the health that is gone, at great expense or possibly with no fix available.

This is like running your car on a low oil level because you don't want to (or thought you couldn't afford to) buy a can of oil, and then when the engine burns out through no-oil friction, calling the repairguy to fix the engine. Often the engine isn't fixable and you either junk the car or buy a new engine. Makes great sense, right? Why would anyone want to suggest that makes for good public policy?

The whole crazy structure assumes one can shop for health care and choose between Walmart, Target, Costco, Safeway, the corner market, the farmer's market, or Whole Foods based on cost or cost/effectiveness. Has anyone seen the price list posted in their medical providers waiting room or outside (like the menu in the window of a restaurant)?

This is economics and health care as only a conservative, ideologue, wingnut could learn to love.

Posted by: JimPortlandOR | Nov 30, 2006 7:54:41 PM

You're right about the out-of-pocket limits. Their averages are $3,172 for individuals and $6,017 for families. The main spending is in the deductible limit, though.

Posted by: Ezra | Nov 30, 2006 8:11:39 PM

I actually currently have an HSA through my employer, on roughly the following thinking: I'm young, I'm healthy, I'm the kind of person insurance makes money on, so why buy it? My median yearly healthcare usage is zero, so who cares?

The satisfaction problem is manyfold: doctors don't understand it, while you can get anything you want there's no collective bargaining benefit so many things are more expensive, and you *feel* every healtcare expenditure in a way you don't on real insurance. That is to say, the way these things save healthcare dollars is to make you wonder, "is my illness *really* three hundred dollars bad?" Yes, it's a better financial deal for median-usage-zero me, so I'll probably keep with it, but I'm sorely tempted to switch to something more likely to make my healthcare happy rather than just profitable.

Posted by: Dennis | Nov 30, 2006 8:17:40 PM

I think HSAs are incredibly stupid for all of those reasons. Everyone who thinks that price transparency is somehow the answer is just deluded. I make a great living figuring out what the health insurance company i work for is going to pay doctors and hospitals. I have the most information about the prices of things, efficiency, and quality. Guess what I do with my HSA money - buy a semi-frivolous lite box for SAD that I never would have without that money. I have an HSA with a 2100 deductible 20% coinsurance and a 2500 OOP max. My employer funds my HSA with $800 and I fund it up to the deductible. That's tax-free money for me and I could give a shit what things cost if i would need serious medical care - even though I could get the most accurate prices of pretty much anyone. Great public policy right? take money out of the risk pool allow me to spend pre tax money. The other problem is that health insurers are too scared shitless of the backlash from doctors and hospitals if you do actually release any useful information. The medical cartel is a huge part of the problem on costs and on quality and efficiency improvement.

Posted by: RebeccaS | Nov 30, 2006 9:55:00 PM

That is to say, the way these things save healthcare dollars is to make you wonder, "is my illness *really* three hundred dollars bad?"

Maybe this is your point, but how would you know? Especially at the point in time where you could, you know, do something about it?

Posted by: Pooh | Nov 30, 2006 10:04:30 PM

All of this discussion skirts the probability that with carte blanche health access without regard or knowledge to costs that the system is over accessed.

Posted by: Fred Jones | Dec 1, 2006 8:49:33 AM

"Probability," Fred? Did you not get your Wheaties with your talking points today? Usually, you are good for at least "near certainties."

Posted by: Pooh | Dec 1, 2006 1:43:36 PM

The RAND folks looked at HSAs and concluded some folks use more care with HSA/HDHP combos and some use less. Glad we got that cleared up . . .

(See my post with a link to the study, at http://healthblawg.typepad.com/healthblawg/2006/10/cdhps_and_hsas_.html )

An imaginary consumer, totally rational in his or her use of an HSA/HDHP combo, who didn't hesitate to actually write a check for health care if it was the right thing to do at the time (cheap/expensive, preventive/curative) might make the plan work, but most real live consumers won't.

The rest, as they say, is commentary.

Posted by: David Harlow | Dec 4, 2006 2:01:30 PM

We are one of the biggest distributor of medical uniforms and we are selling our scrubs for only $9.99 per set. Just go to http://www.cheapscrubset.com

Posted by: jose gomez | Sep 24, 2007 8:54:01 PM

HP F4098A
HP F4809A
HP F4812A
HP PP2182D
Pavilion ZV6000
HP F2024B
Pavilion 5000
Pavilion ZV5000
HP PP2812L
Pavilion ZD7000
Pavilion ZD8000
Pavilion DV1000
Pavilion ZT3000
Compaq Presario 2500
Compaq PP2200
Presario NX9000
Compaq Presario R3000

Posted by: batteries | Oct 20, 2007 9:27:03 PM

The comments to this entry are closed.