« Should Bernanke be reappointed? | Main | Totally false and ridiculous claims about Chinese bubbles »
The Uninsured: Adverse Selection Problem or Distribution Problem?
In his recent post on health care and insurance Paul Krugman writes:
[Insurance companies] try to avoid covering people who are actually likely to need care.
If insurance companies do avoid covering people who are "likely to need care," this suggests that the uninsured are unhealthy. But 60% of the uninsured are in excellent health (Table 10) (In fact, overall the uninsured are only slightly less healthy than the insured).
To be sure, this doesn't mean that being uninsured is not a problem but, contra Paul, it does mean that insurance companies would be willing to cover most of the uninsured at the same rates as the insured if the uninsured could or would pay those rates. In Paul's story there is a market failure, in the latter story health insurance is expensive and some people don't buy it. The difference matters because the wrong diagnosis will almost surely lead to the wrong treatment.
Addendum: McArdle nicely takes the time to follow the logic.
Posted by Alex Tabarrok on July 27, 2009 at 07:20 AM | Permalink
Comments
for there to be a "market success" wouldn't you need both the insurance companies covering people at existing rates PLUS some ability for people to see and understand the basis for the cost of their insurance? and for that, people would have to assess their health care needs properly _in advance_?
Posted by: babar at Jul 27, 2009 7:31:34 AM
'If insurance companies do avoid covering people who are "likely to need care," this suggests that the uninsured are unhealthy.'
No it does not. It suggest that the unhealthy are uninsured. Not the other way around.
Imagine a market where 18% are 'superhealthy' (do not need care and hence will not buy insurance), 70% are normal and will buy insurance and 12% are unhealthy and will be excluded from the market. Voila, 60% of uninsured are in excellent health.
Posted by: Morten at Jul 27, 2009 7:46:41 AM
I have not read Krugman enough to say that he always preaches the truthy liberal preconceptions to his choir, but I haven't seen otherwise. Keep hammering him on these points Alex. Bravo.
And again, the unhealthy don't need insurance. Insurance is for the pooling of risk. There is no risk the unhealthy will become unhealthy. They are unhealthy. They don't need insurance, they need wealth, income, or charity.
Posted by: Andrew at Jul 27, 2009 7:48:18 AM
That would be true if the sick had the same demand for health insurance as the healthy. However if sick people obtain greater utility from insurance but are equally likely to purchase it that suggests they are being priced out. This could easily be due to the negative correlation between health and income rather than insurance company discrimination but it's still there. I don't think those stats necessarily back up your point although they don't by themselves back up either argument. Either way it doesn't seem like market failure. Functional insurance markets will absolutely price out very sick people. Letting them die isn't really a failure of the market.
Posted by: Ryan at Jul 27, 2009 8:01:04 AM
Morten, that's why I said "suggests" rather than "implies." In anycase, key point is that insurance companies would be willing to cover most of the uninsured at the same rate as the insured.
Posted by: Alex Tabarrok at Jul 27, 2009 8:11:07 AM
Alex, couldn't both stories be true? What I'd like to know is what percentage of the uninsured population is uninsured by choice. Moreover, the fact remains that our health care system (if there is anything systematic about it) is extremely expensive compared to others for roughly equivalent health outcomes.
Posted by: dan cole at Jul 27, 2009 8:21:29 AM
ZOMG! I love economists. "If insurance companies do avoid covering people who are "likely to need care," this suggests that the uninsured are unhealthy." Dude, insurance companies spend huge amounts trying to avoid covering people who are likely to need care. It's called underwriting, or in more extreme forms, rescission, and it is not secret. They spend a lot of money on it.
At its most benign, underwriting just means putting a price on the policy that is commensurate with you risk. So if you will have a $100,000 claim next year, the insurance company will set your premium at $100,000 or deny coverage. At its worst it is rescission of a medical policy with a large claim for specious reasons, such as a doctor's visit for acne years ago.
I would suggest that every two years economists need to take a reality check. That could help prevent embarrassing posts like this one.
Posted by: es32 at Jul 27, 2009 8:25:01 AM
People tend to get covered through their employers. Employer coverage usually covers all employees. In this case, insurance companies can't choose who to cover. They can change pricing and payouts.
Posted by: fusion at Jul 27, 2009 8:29:58 AM
Alex, do you agree or disagree with Krugman's claim that insurance companies use a lot of resources trying to avoid covering the sick? If that's true, then it's a problem whether or not the sick end up getting covered. The insurance companies could be competing over who ends up covering the sick.
Also, it's not clear that the difference is all that slight. Take a look at Table 1. Table 10 lets you see who's uninsured, but Krugman's claim is that this is a problem for sick people who don't get care, not that it's the main explanation for why people are uninsured, so I think it's more relevant to see what sort of insurance the sick have (which is in Table 1). 21% of those in fair/poor health are uninsured, compared with only 15% of those in excellent/very good health (and you'd expect those in poor health to be more motivated to get insurance). Further, those in fair/poor health are much more likely to receive public health insurance (Medicaid or other) than those in excellent/very good health (39% vs. 13%). They are also less likely to have individual private insurance (3.7% vs. 6.1%).
Posted by: Dan at Jul 27, 2009 8:40:09 AM
The people who saw a market failure in housing for the poor gave us public housing projects. How did that work out?
Now they have turned their attention to health care, only they want to place everyone in their system. Why do some think they will make better doctors then landlords?
The biggest group that lacks medical insurance? Families of illegal immigrants. Why aren't politicians targeting this group in their public relations drive for national health insurance? Why do they object to just opening free clinics in high concentration communities to serve this population. Texas has worked on this. Is this a market failure, a moral failure, or a lack of political courage?
Posted by: DanC at Jul 27, 2009 8:41:13 AM
I think your evidence might suggest that the answer to the question in your title is, "yes." Some theories point out that the uninsured should be less healthy, i.e. that the reason that they are uninsured is because they are denied coverage due to the fact that they are unhealthy. Other theories suggest that it is unhealthy people who demand insurance on the individual market, thus leaving those who are uninsured as those who are more healthy. The distributional theory suggests that they are just as healthy, but that the uninsured are just more budget-constrained. The figures you cite don't sort out these effects.
There is a White House study out today that indicates higher per-person prices for small businesses than large ones. There might be some economies of scale here, but could something else be at work?
Posted by: Jim B. at Jul 27, 2009 8:44:49 AM
Could something else be at work?
Yes, the basic insurance concept of pooling.
The larger the group being placed in the pool the closer it is to a normal population.
Small firms pay higher rates because they have a smaller population to be pooled.
At my wife's small firm the owner had a major heart attack and it forced everyone's health insurance premiums higher.
Posted by: spencer at Jul 27, 2009 8:52:48 AM
Here's another way to slice the data (doing some calculations based on Table 1). Of those who get private insurance (either through their employer or individually), 5% are in fair/poor health and 76% are in excellent/very good health (the numbers are almost the same for the 2 sources of private insurance). Among the uninsured, 10% are in fair/poor health and 60% are in excellent/very good health. So I doubt that "insurance companies would be willing to cover most of the uninsured at the same rates as the insured."
There are a lot of sick people with public health coverage, which is why the gap in Table 10 doesn't seem so big. On Medicaid, 17% are in fair/poor health and 57% are in excellent/very good health. On other public health coverage, 37% are in fair/poor health and 39% are in excellent/very good health.
Posted by: Dan at Jul 27, 2009 9:04:26 AM
If insurance companies do avoid covering people who are "likely to need care," this suggests that the uninsured are unhealthy.
No it doesn't, it suggests that the unhealthy are uninsured (i.e. 4.6/21.9 = 21%, vs. 15% for ex/vg). If insurance companies tried to avoid covering left-handed people, that wouldn't mean the uninsured are left-handed. Seriously, this is GRE level stuff.
Posted by: ogmb at Jul 27, 2009 9:06:37 AM
There is no risk the unhealthy will become unhealthy. They are unhealthy. They don't need insurance, they need wealth, income, or charity.
Or insurance they bought when healthy, which cannot now be rescinded when the insurance company has lost its bet.
Posted by: Bernard Yomtov at Jul 27, 2009 9:09:50 AM
What Krugman - and most economists - seem to ignore is that 35 states already have laws which force insurors to cover the high risk population. High risk pools are available for almost everyone who has been turned down by health insurance companies. Premiums for high risk pool insurees vary from 150% to 250% of the average rate for non-group policy holders. These are below what a free market would charge, of course. Normal insurees are paying higher premiums and subsidizing the high risk pools.
In a handful of states, high risk pools are capped, and potential insurees must wait for coverage. But such waiting lists are the exception rather than the rule.
I do not advocate such interference in insurance markets. However, I find high risk pools to be much less intrusive than the outrageous proposals being rammed through Congress right now.
Posted by: John Dewey at Jul 27, 2009 9:17:53 AM
I would say that cute little statistic isn't significant unless you also include the percentage of people in excellent health who do have insurance because a discrepancy could be used to support what Krugman said right? What a joke.
Posted by: bedmondson at Jul 27, 2009 9:42:04 AM
I have read in various sources that a
proportion of the young who are in good health
opt out of insurance because of the expense.
I wish I knew the answer to Dan Cole's question
about the exact figure. I read an article in
the Washington Post today, which said that the
uninsured get 50-70% of the medical care the
insured get: probably through public clinics
and other agencies. I do not know its quality.
People with a pre-existing condition either have
to get group insurance through an employer or do
without, the individual option not being open to
them.
Posted by: Candadai Tirumalai at Jul 27, 2009 9:53:31 AM
N.C. high-risk pool an option for many seeking health insurance
"The plan covers three groups of people: Those who cannot get group coverage and face higher premiums because of pre-existing conditions, those who have run out of COBRA coverage after leaving a job and trade-displaced workers who have lost their jobs because of globalization."
I'd rather see 50 states trying to independently work out coverage for their uninsured than see the creation of a new Federal behemoth.
Posted by: John Dewey at Jul 27, 2009 9:55:03 AM
Ahh, nothing like statistics to cloud the issue.
The bottom line is everyone needs to have health care. The days of personal bankruptcies due to medical bills needs to end. All the statistics in the world can't change the fact that this happens today in America.
This care isn't free, so everyone has to pay. Payroll deduction, income tax, whatever. Unless we as a society are ready to close the hospital doors on those who chose not to buy insurance (and I am not) than you don't get a choice. Insurance is a risk pool, and it only really works if everyone is in.
A lot of this discussion is just partisan noise that distracts from the goals above.
Posted by: Jim at Jul 27, 2009 9:55:48 AM
'key point is that insurance companies would be willing to cover most of the uninsured at the same rate as the insured' - if they insured everyone en masse, that is. But you are not making a point about what would happen if there was mandatory universal coverage here; your data doesn't say anything as to whether there is a large group of people that insurance companies won't cover. There definitely are lots of people who choose not to insure, but there are likely to be lots of people who are denied insurance *at the going rates* too.
1. Don't treat health status as a binary variable just because this report does! You can be many different types of unhealthy.
2. Also, does health status refer to *current* health status or health status when taking up insurance?
Your points are almost invariably insightful even though often counterintuitive, but you are spectacularly wrong here.
Posted by: Datacharmer at Jul 27, 2009 10:11:12 AM
jim: "The days of personal bankruptcies due to medical bills needs to end."
Some choose not to manage their lives. That's no reason the rest of us have to pay for the consequences of their mismanagement. If you believe others are entitled to your earned income, fine. Go find an uninsured person and pay his medical bills. But don't impose your values on me and have the government take my money.
Posted by: John Dewey at Jul 27, 2009 10:13:42 AM
Candadai Tirumalai: "People with a pre-existing condition either have to get group insurance through an employer or do without, the individual option not being open to them."
That's not correct. 35 states have established high risk pools which force insurors to provide coverage to such persons.
Posted by: John Dewey at Jul 27, 2009 10:16:32 AM
Insurance isn't health care. Insurance is a financial instrument designed to protect people from the unaffordable costs of an undesirable, uncommon, and unpredictable event. Insurance is not a mechanism to pre-pay for or shift the cost of something desireable, common, and (sometimes) predictable. What the nation is struggling with is the cost of health care. Why we blame insurance companies for this is a matter of poorly educated journalists and opportunistic politicians.
It is simply the case that as a society, we have more health care technology 'available' than we have resources to pay for everyone who wants it. Imagine if everyone, in 1970, had wanted/needed a computer with the power of today's typical PC (ie, a Cray). The dilemma we face is that we cannot stomach people dying because they can't afford a medical procedure that, but for the cost, would otherwise be available.
Who pays isn't the problem. Limited resources is. But Congress is almost exclusively focused on the 'who pays' question. The 'market' would have sorted this problem out long ago but for government subsidizing Medicare and Medicaid. That has encouraged the creation of relatively expensive technologies knowing that the government was picking up the tab, and therefore cost was not a consideration. Putting everyone in a system backstopped by the government will only make the problem worse until government is forced to ration by means other than price.
If people paid for their own 'routine' health care, and only bought health insurance for the truly insurable risks (including 'health status' insurance), the market would be signaling providers to lower costs and only extend life to the extent that people could afford it. As long as 'someone else' is picking up the tab for extending life, the market is being told that cost is no object and we can't be surprised that we get very expensive, unaffordable, drugs, devices, and care.
PS. Who knows of a good study on correlating American's relatively poor health with the nation's health expenditures, relative to other countries that spend less, but are healthier (eg, less weight-related illnesses).
Posted by: Richard at Jul 27, 2009 10:19:13 AM
Why did you pick this paragraph out of Krugman's article? Granted, if you disprove him on that argument, then his whole post is not much better than what an 8 year-old suffering head trauma could whip up. He has other paragraphs, such as this one:
There are two strongly distinctive aspects of health care. One is that you don’t know when or whether you’ll need care — but if you do, the care can be extremely expensive. The big bucks are in triple coronary bypass surgery, not routine visits to the doctor’s office; and very, very few people can afford to pay major medical costs out of pocket.
Has he never heard of fire, flood, or auto insurance? This is just a stunningly stupid paragraph for a guy with a Nobel Prize. He goes on to show he's never read or forgets his Adam Smith. It's almost like a free-market economist is ghost writing to set up the perfect straw man.
Posted by: 8 at Jul 27, 2009 10:29:28 AM
