Competition and Concentration in Health Insurance

Many people have bandied about numbers suggesting that the market for health insurance is highly concentrated.  Here is the President:

Consumers do better when there is choice and competition. Unfortunately, in 34 states, 75% of the insurance market is controlled by five or fewer companies. In Alabama, almost 90% is controlled by just one company….

But these statistics only include people insured by "insurance companies" even though nationally just over half of all employees get their health insurance from a firm that self-insures.  In other words, as John Lott points out, over half of the market for insurance is being left out of these concentration statistics.

Since about half of employees are insured by a self-insurer, concentration statistics–as typically presented –should be cut roughly in half (precise numbers vary by state).  Firms that self-insure typically outsource benefits management and claim
administration to highly competitive third party administrators.  A key fact according to this paper (which is outdated although I wouldn't expect the basic finding to have changed) is that the populations served, the benefits paid and the premiums paid are about the same for firms that self-insure and firms that buy insurance from a health insurance company.  Thus, concentration among that part of the market served by health insurance firms appears to be well disciplined by the larger market for self-insurance.


I would like to add that in many states, the largest insurer is a non-profit (not counting employer self-insuring). For example, in Alabama, Arkansas, Maryland, Massachusetts, Michigan, Rhode Island, and Vermont (amongst others), a non-profit company controls at least half of the market. So even if those markets are concentrated, it is by a firm that is not controlled by the evil profit motive.

So the concentration is lower than indicated and non-profits control much of them. The claims are all very misleading.

Your logic is flawed. I can't go and buy insurance from a self-insurer, so they are not in my consideration set.

All this says is, that with any monopolized market, the consumer always has the option of producing the service himself, even if this is grossly inefficient. The stronger the monopoly, the higher the price, and the more self-production will occur. Self-production will start with the subset of consumers who are relatively efficient at this, leaving the remainder of consumers facing the monopolist.

No, the logic is not flawed. There are all kinds of quasi-monopoly situations, in fact you'd be hard-pressed to name one that isn't.

The logic that a government monopoly combined with illegalization of self-insurance will increase choices is the flawed logic. It is so obviously flawed as to be a lie. Obama is a liar.

Jon: Lot's of people make home movies and even have their kids put on hilarious, heart-warming skits, so the local cable company doesn't have a monopoly as long as you include this self-production.

What's funny is that Obama is a supporter of single payer, because he thinks that things would be better if the government used monopsony bargaining power to lower prices in health care. Why he thinks that Blue Cross and Blue Shield of Alabama, a nonprofit with 90% of the market (as he defines it), would do a better job if they faced more competition is a mystery. Maybe the problem is that they're nonprofit?

I think the self-insures line is both accurate and misleading. My understanding is that firms that self-insure still use the networks and pricing established by the insurance companies. To the extent that Obama's complaint is that the insurance comapnies aren't driving down prices, firms that self-insure aren't relevant, or aren't relevant to the extent they piggy-back on pricing.

It's not just companies, but individuals who self-insure.

In fact, it's interesting how millions of people with pre-existing conditions (like victims of domestic violence) are among the consumers who are most aggressive about "self-insuring".

They must be the major insurance companies fiercest competitors!


Libertarians don't favor large corporations; modern liberals do however. In other words, the sorts of regulatory polices that modern liberals push favor large corporations.

a student of economics,

Lot's of people make home movies and even have their kids put on hilarious, heart-warming skits, so the local cable company doesn't have a monopoly as long as you include this self-production.

Well, and I realize you are trying to analogize here, but cable isn't the only option, Sat. TV is another. And of course as a practical matter cable TV isn't simply competing against Sat. TV or home movies; they are competing against a vast array of entrainment choices that range from what is on the internets to to Netflix to live performances to running in the local park.

If they wanted to really address concentration in the medical services market, they'd be talking about this:

The High Concentration of U.S. Health Care Expenditures
Research in Action, Issue 19
Half of the population spends little or nothing on health care, while 5 percent of the population spends almost half of the total amount. Examining the distribution of health care expenses among the U.S. population sheds light on areas where changes in policy might bring about the greatest savings.

Are they talking about this? Hell no. And they don't want people talking about it because it undermines their moralizing on the issue. They want you to believe that all the little children aren't getting their shots.

So, the only problem is the irresponsibly uninsured. How could you deal with the uninsured? You could assess them a fee when the get service. Give them what amounts to a government loan against all future entitlements such as Social Security and medicare. There are of course issues here, but I spent about 5 minutes coming up with that, where the hell are the "experts?" Again, they are not so stupid that they don't know what the issues are.

It's useful to point out that more concentration may occur in some states because each state has its own licensing and regulations for insurance companies to sell policies within the state. Thus, some states may be less attractive to insurance firms. In addition, federal law does not allow insurance to be purchased out of state. McCarran-Ferguson Act (1945),protects state insurance firms from interstate competition.

Of course, if we were really worried about concentration in insurance markets, we would allow cross-state selling of insurance which would not only reduce concentrated power by insurance companies but also state insurance bureaucracies.

Cash: To me, being a non-profit means $ that would otherwise go to shareholders gets drained off by management in the form of cushy workplace arrangements.

Which is to say, it's nothing like a govt agency staffed by civil servants.

Actually, it sounds to me pretty much exactly like a government agency once you replace the word 'shareholders' with 'taxpayers'.

@John Thacker
"Wal-Mart has done wonders to improve the health insurance market, if you count lower prices on generics prescription drugs. (Not that they're the only large retailer with such a program.)"

Yah, the $4 generics and $20 flu shots are GREAT.

This non-profit for profit stuff is, pardon the expression, bs too.

Organizations with market power and with the ability to extract profit position can distribute the profit to shareholders, OR, if they are non-profit, can lavish it on their executives and managers. Although there is literature on the behaviour of for profit and non-profit hospitals in the exercise of market power, the consensus is that it does make a difference in terms of price/cost performance. But, as a lawyer, you still make the argument anyway and hope no one is doing their homework.

Andrew observes:
"AIG can fail and it would be painful whether or not the government bails them out, but at least they can fail if they do a terrible job. The government cannot fail without destroying many of us, that is why big corporations are better than the government"

May I introduce you to fascism, Andrew?

Wal-Mart would actually probably really like the Baucus plan, with medicaid coverage up to 133% of income. If an employee works 35 hr/week and 50 wk/year, they make $12,583 at federal minimum wage. Which means a single earner with 1 or more dependent would be covered by medicaid, and Wal-Mart wouldn't be paying for it with the Baucus version of the employer mandate.

"Of course, if we were really worried about concentration in insurance markets, we would allow cross-state selling of insurance which would not only reduce concentrated power by insurance companies but also state insurance bureaucracies."

Sure, if you want widespread regulatory arbitrage and the mockery of state's rights. Unless, of course, you're also suggesting to go along with this federal regulations under Art. 1 Sec. 8's commerce clause that sets a nationwide floor on what health insurance has to cover. Which is similar to what's being proposed, except with the addition of an open marketplace that improves the customers access to information and choice. This is one issue where both liberals and conservatives agree, they just disagree on the solution. In this case the liberal side is more correct, and the major objection is the fact that the federal government is involved.

John Dewey @ 11:21am

That's exactly what I was thinking.

So why isn't Obama all over the price discrimination against those paying cash or with small providers?

(chirp chirp chirp)

The point is not that there isn't "market power." It is that Obama conveniently dismisses fully half the insurance providers.

It would be like taking the 80/20 rule and calling it the 20 rule.

Self-insuring is like vertical integration - instead of buying a product on the market, you make the product yourself. As insurance companies charge higher prices, it becomes economical for smaller firms to self-insure. One of the problem is that state regulations place a lot of restrictions on insurance providers, so smaller firms face a greater percentage of overhead expenses to comply. This may not figure in to antitrust market considerations, but it nonetheless serves as some kind of check on abuse of market power.

I do not deny that an analysis by the FTC or DOJ would look at the market power of a non-profit. However, when the President claims that it is the profit motive that is driving up costs, that earning money for shareholders is the problem, it becomes rather relevant that non-profits have a dominant market position in many states.

The FTC/DOJ would also look at potential market entrants. I would imagine that at least one big national insurance company would qualify for a given market. Additionally, market power alone means nothing.

"Even if the concentration numbers are lower, the more important metric of "plans available to consumers" is the same: for the overwhelming majority of us, our "choices" are:"

1) the plan offered by our employer;
2) nothing.

Not true.

Part of why the insurance market is top-heavy is not so much that the choices are limited, but that we consumers don't want to think about it and go with the lowest common denominator while other options are available.


As those of us who have either defended hmos, or who sell insurance to large employers have said,
the small employer and individual market is where the problem is. These folks can't self insure. If you go back and look at the tables in the document linked to the original document, you will see that small employers (and for that matter individuals) of less than 200 employees do not self insure. Some of this may also depend on whether provider discount networks have developed in a state (and some southerns states do not have discount networks because managed care is anathema to them and the docs and the hospitals have done all they could to block manage care networks).

So, you question was: why don't other health insurance companies enter. At first blush, you'd say, hey, a big insurance company in one state, why doesn't he take a toehold in another--he's big outside, he would have more assets than the incumbent. That's true, and insurance regulation isn't all that costly, and the minimum elements of insurance coverage may be defined by state law, in other words, there is a pretty uniform stardard which you can add bells and whistles to?

So, why not potential entry a limit on power? Well, the larger incumbent plans have done a combinatin of things: 1) negotiated discounts with docs and hospitals based on their market share; 2) entered into most favored nation clauses or meeting competition clauses with hospitals or plans--so if a hospital were to offer a lower price, it would be offered to the incumbent, but, since the incumbent has a larger market share, the cost to the hospital wouldn't make it likely--it would lose more than it would gain with the prospect of uncertain increase volume in exchange for the certain loss of revenue from the incumbent. So entrants typically pay house rates, and incumbents get a deal. But, there's more: hospitals don't want managed care or anyone else in that market and they are very happy with this arrangement of a passive bluecross.

So, what you really need--and I know you don't want to hear this--is someone with a book of business who can come to the hospital and say: you know, I want that same deal as BCBS, and I want you to improve outcomes, and here are some quality protocols I want you to consider (if you don't consider them, you have to have a reason with credible research because we have credible research supporting these Mayo or other guidelines). Now, who do you think could come to the table with this book of business? This book of business that could be used to promote entry and break up the tight block of alliances between hospitals and the incumbent carrier? Hmm. I'll let you think about that for a minute while I mention one other thing from business.

Pepsi and Coke were getting, what they thought, were bad deals from the High Fructose Corn Syrup Folks. HCFC is a commodity, standardized by chemical composition. Yet, Coke and Pepsi knew, probably from engineering studies, that they were paying a lot for the product.

So, they announced they would move some business to anyone who qualified as a new entrant, and if they had their plant inspected, they would move some business. Guess what: someone entered. And, the plants that told Coke or Pepsi to stick it where soon scrambling to sell to them.

Same here. The Fed, for some class of customers, is footing the bill. Like any business buyer it wants competition. You should too. No one is better off without competition. Some of these markets (on the hospital and the insurance side) are lazy and will rip you off.

Ultimately, if we do get some form of national competition in the procurement/insurance market, we can start improving our healthcare system. Nothing opens a closed mind than the reward of payment for performance and quality. We'll never get they if we embrace the past.

I beg to differ with the commenter who said your business needs to be "larger" for self-insurance to work. I served on the board of a non-profit with about 90 employees within the state and we routinely shopped our group to Blue Cross, self-insurance packagers, and for-profit insurers. They're close to being perfect substitutes. Self-insurers not only use "outsource benefits management and claim administration to highly competitive third party administrators", what's more they're not really self-insured, it's coinsurance. You buy stop-loss insurance in a highly competitive market as well.

Many small businesses are left out in the cold -- that is, pay high premiums -- for the same reasons individuals are, whether you subscribe to the adverse selection argument, the argument that insurance companies merely know who is likely to be high cost and set the premium accordingly, or the argument that part of insurers costs to be covered in a competitive market are the costs of screening low from high cost. Larger companies (and they don't have to be much above 75 employees) create a pool sufficient for these issues to disappear.

The data in the paper cited is outdated... the recent data is that the largest companies have premiums, on average, 18% lower than small businesses.

It is also why big insurers have vehemently opposed Association Health Plans/ Small Business Health Plans.
Even the 2005 Senate bill was modeled to result in a 12% decrease in small business premiums (done by Mercer).

btw, glad the experts are finally coming out of the woodwork. Firstly, I can shut up. Secondly, it's amazing that on almost every facet of the issue the conventional wisdom that is used to push the politics is 180 degrees wrong.

Just for example, people really think we can cut medical costs by reducing the supply of doctors. Astounding.

Bill, point taken that my sense of the market is more anecdotal than systematic and thank you for the pointer to the DOJ. I'll take a look.

As to another related thread of conversation, there's another reason that self insurance has advantages -- it gets around state regulations. The firm is merely the agent of the employee seeking insurance. In other words, the employee in a self insured firm is getting around state regs which purport to protect the public but which do not necessarily do so. I'm not sure why the states rights argument is compelling if one form of insurance (self insurance) can get around state regs and others cannot. We've already crossed the boundary where regulation of insurance was left to the states. Wouldn't it be nice to be in the realm where states compete with each other for in terms insurance regs, just as they compete with each other in terms of creating an environment for free enterprise.

Finally, as Wal Mart has been raised a few times, remember one reason for it might be interested in expansion of federal regulation of health insurance: raising the costs of its smaller rivals.

"The good part of Obama's plan is that it permits national plans to be offered...something to compete with local, state protected plans."

Which plan/bill does this??? URL please???

@ Bill at Sep 17, 2009 6:07:10 PM,

Took a look at DOJ. I can see the DOJ is concerned about insurance mergers (aside: while rightly looking at it on a geographic basis), and that it does not seem to consider the possibility of self insurance -- although I didn't enough time surfing to say so definitively.

DOJ is concerned as well with mergers of HMOs and other doctor groups -- which you note. But I believe this throws a red herring into Alex's post which is about insurance concentration, not provider concentration. Unless one wants to argue TPAs cannot bargain with doctors as well as traditional insurers can.

"Self-insurance" isn't insurance, it's just a fancy term for putting your money in a savings account, or piggy bank, for emergencies. I'd like to see what percentage of these self-insurers have coverage for catastrophic illnesses.

Finally, as Wal Mart has been raised a few times, remember one reason for it might be interested in expansion of federal regulation of health insurance: raising the costs of its smaller rivals.

Comments for this post are closed