Monday, August 24, 2009

More Inconvenient Health Care Facts: The Role of Self Insured Employers

I have previously mentioned that Dartmouth College was self-insured in regard to health care costs of its employees, and that this creates a situation wherein the College not only has the incentive to manage its employees' health costs but also has tremendous ability to do so.

In discussing this with people, I am struck by how many did not realize the extent of self insuring employers. So here is a document that gives the facts: Health Plan Differences: Fully-Insured vs. Self-Insured, from the Employee Benefit Research Institute.

As I suspected (ah, theory does help one predict accurately!) the extent of self insurance among employers is quite great. Overall, 55% of workers "with health insurance" were covered by a self-insuring employer in 2008. Even more telling, 89% of workers in firms with more than 5000 employees were covered through a self-insurance plan.

Most companies that self insure will still use a Third Party Administrator (TPA) -- Dartmouth uses Anthem BC/BS -- to handle the administration of the health plan. Besides processing all the paper, this also includes the important role of negotiating rates with medical service providers. Self insurance simply means that the company will pay all of its employees' health care bills -- they just get passed on by the TPA. (Most employees, however, probably think that they are being insured by the TPA, and when there are issues with nonpayment, etc., they will blame the TPA.)

How does the great extent of self insurance fit into everyone's perception of the greedy, rapacious role of insurance companies? They are not even insuring most workers!! And, the Third Party Administrator market is a very competitive one. A company can request bids from many TPAs and take the best one. And the employer can design its health care plans as best fit its employees, and then just tell the TPA to administer it. That is, as I have repeatedly argued, the employers have all the incentive and ability to manage their health care cost. The problem is NOT with insurance companies. This is one of the myths being consistently pushed by those who want reform at any cost.

Sunday, August 23, 2009

More Health Care Facts

The columnist Steve Chapman at the Chicago Tribune has written a couple great columns on dispelling some health care myths. The latest is here. This column visits some of the infant mortality data that I discussed below. One good quote is:
Nicholas Eberstadt, a scholar at the American Enterprise Institute in Washington, also attributes the gap largely to conduct. Comparing white Americans to Norwegians in his 1995 book, "The Tyranny of Numbers," Eberstadt concluded that "white America's higher rates of infant mortality are explained not by poverty (as conventionally construed) or by medical care but rather by the habits, actions and indeed lifestyles of a critical portion of its parents."



When I was discussing infant mortality the other day, some folks were all too ready to say that the US health care system is also to blame for the tendency in the US to have low birth weight babies. If only those teen-age moms had seen a primary care physician, they said, they would have stopped smoking and using coke, and -- voila! -- we would have had a nice strapping 8 lb. baby. Ah, one has to marvel at the optimistic trusting innocence of the liberal mind.

I guess also that Steve Chapman is likely to get branded as one of these right-wing maniacs who are spreading misinformation and disrailing the high speed locomotive known as health care reform.

Health Insurance Fallacies

The proponents of health care reform like to argue that opponents are spreading misinformation to push their views.

I need to point out the one fallacy that I consistently hear from reform proponents, and that has sunk in due to its relentless repetition. I have had colleagues fall into the trap and many students as well, even those who have had several econ courses.

The fallacy is illustrated well by this quote from an AP story written by Ricardo Alonso-Zaldivar:

Proponents of a government plan say it could restore a competitive balance and lead to lower costs. For one thing, it wouldn't have to turn a profit.


Wouldn't have to turn a profit!! There is almost no economic nonsense that upsets me more than this kind of thinking -- that when we buy something from a company, we are paying an unnecessary "profit". In this make-believe world, profit is a "cost" that has to be paid, but only if we are buying from a private company. If only that private company were substituted for by a nonprofit organization or a government agency, then we could get the same service but at lower total cost. Magic!

Maybe it is the language of "for-profit" and "not-for-profit" that leads to this mistaken conclusion.

If only it were so simple to avoid an unnecessary cost! If it were, every single company in the world would fully vertically integrate, making all of its inputs itself -- why buy anything from another company and pay them an unnecessary profit? (Of course, this flies in the face of another trend that upsets many, that of outsourcing.)

And closer to home, nobody would lease a car -- why pay a profit to the owner, a bank -- and nobody would ever rent a house -- why pay a profit to the landlord?

The correct lesson from the most important part of any economics course, that on the organization and performance of competitive markets, has tragically failed to sink in. In a competitive market, any excess profit will get competed away. The only profit is a normal -- that is, the minimum acceptable -- return on invested capital. For example, if an insurance business requires $1 billion of invested capital, then at a 10% normal return, the level of profit will be $100 million. You should think of that $100 million not as a profit, but as a cost -- the cost of having that $1 billion invested in the insurance business. If the insurance business yields a profit of $125 million, then more capital and resources will enter that business, driving the return back down to normal levels. That is, there would have been in this case $25 million of "economic profit" and competition will drive that away.

A not for profit firm, or a government agency, cannot avoid incurring this cost of a normal return on invested capital. There is absolutely no sensible argument that a not for profit would not need the same amount of invested capital (buildings, equipment, cash on hand to pay employees and claims, etc.). And with the same amount of invested capital comes the same cost associated with its deployment. (In fact, with the public option being discussed in health care, the US government would give the new public organization an infusion of capital to get started. Do we think that capital comes free?)

One simply cannot argue that a not-for-profit or government agency can produce at lower cost because they can avoid paying a profit. The only way to argue that a not-for-profit or government agency can do something for less cost than a private entity has to rely on an ability to actually run the business more efficiently (fat chance!) or truly avoid some real cost (I will give you one idea here -- WHO DOES NOT PAY TAXES!)

This argument does of course rely on competition driving the return on capital to normal levels so it does not apply to an industry with significant monopoly power (if a company faces a monopolistic supplier of an input, then vertical integration can make sense). I am in no way prepared to accept that the insurance business is now or would be in the future characterized by enough market power to prevent the inexorable forces of economic competition from driving expected profit to normal levels, the arguments of folks such as Alonso-Zaldivar notwithstanding. My next post is going to revisit the issue of self insurance, which will raise even more questions about excess insurance company profit and the claimed lack of competition.

Friday, August 21, 2009

Google Voice, Apple and ATT

There is an increasing buzz around Apple's "rejection" of the Google Voice application for the iPhone. Andy Kessler, writing in the Wall Street Journal, lays out the usual case against Apple. Apple meanwhile has denied that it even rejected the app -- claiming it is just studying the issue -- and denies talking to its partner AT&T about it. See this article by Saul Hansell in the NYT, for instance.

There are many interesting aspects to this case, all worthy of careful thought.

One of my first reactions is: history does repeat itself. Many years ago, a startup long distance company called MCI forced AT&T, then the sole long distance carrier in the US, to allow MCI access to its trunk lines. MCI had a business model of leasing lines from AT&T and selling long distance service to business customers at a significantly lower price than AT&T. At least initially, AT&T did not want to lease lines to entrants like MCI, for obvious reasons.

This case also smells a bit like the big antitrust case against Microsoft, particularly in regard to the claim that Microsoft forced computer sellers to put only Internet Explorer on their desktops.

So the claim against Apple/AT&T is essentially one of creating barriers to entry through exclusion. (Besides the antitrust issues here, there are the usual "free internet" folks who hate to see any kind of proprietary/exclusionary activity on the internet.)

There are of course some serious differences in these situations, and I certainly don't agree with all of the antitrust claims against Microsoft anyway.

Both Apple and AT&T potentially have something to lose from the Google Voice app. (The app is pretty neat, one has to admit. It consolidates all one's phone numbers in one Google phone number, allows free domestic calls and text messages, cheap international calls, and some other very nice service features. It is available as a standalone service; the current controversy concerns only the offering of Google Voice via an iPhone application.) Apple may have a concern about Google gaining a foothold in the mobile phone market -- similar to claims that Microsoft by excluding Netscape was trying to protect its operating system market. AT&T's concerns are more direct: if its customers use Google Voice for texts and calls, it loses revenues immediately.

Apple/AT&T have several defenses that are very credible. Foremost would be the simple argument that Google is freeriding. Apple and AT&T have invested to build the iPhone business, and are continuing to invest to make it better (such as -- listen up, AT&T -- getting 3G service in places like Hanover). In order to recoup those costs, AT&T has to charge somehow. Demand elasticities make the pricing of things like international calls, and maybe text messaging, the best place to extract revenues. Along comes Google, essentially saying, "thank you very much for building this platform and discovering all these customers; now I will simply go in-between you and the customer and offer them a better deal." Google of course does not need to recover the costs of that big national wireless network of AT&T's, which is one of the major factors giving such a large customer base to the iPhone.

Imagine Coca Cola building up a customer base who love the taste of its product, and acquiring a fully owned/integrated bottler network to distribute its product. How would we feel in forcing Coke to let a coke imitator use the Coke bottling network to distribute its free-riding product to Coke's customers?

Apple and AT&T also do not have the kind of market power that one might have ascribed to AT&T of old and even Microsoft. There are numerous competitors to the iPhone, and there are good competitors for the AT&T network as well. Just look at RIM and Verizon. What we have going on is really a very healthy competition between different mobile phone platforms. Apple, true to form, is going with a fully integrated model (well, AT&T is not owned by Apple but it is a pretty tight partnership). RIM/Blackberry and others are more like the Microsoft/Windows model: don't worry so much about a fully integrated system, instead let competition at each stage determine the best parts and "hope" that they work together well. Why would we want the government to get involved in this battle that should be determined by the consumer? If consumers don't like the Apple/AT&T exclusionary model, then let them eat cake...oops, sorry, I mean let them buy a Blackberry and sign up with Verizon.

I do think though that AT&T in particular needs to be careful about its pricing of some services. It's international roaming rates are horrendously high, for example. One of the beauties of the iPhone/AT&T deal is unlimited internet access and virtually unlimited domestic calls and text messages (with the right plan). For some customers, however, I suspect the global charges are going to be the margin where purchase decisions are made.

Thursday, August 20, 2009

Health Care Thoughts

It is most entertaining to talk about health care with intelligent people. The differences in beliefs, and in philosophical orientation, are just amazing. Some of the difference in belief can be eliminated with data. The philosophical differences are harder to get rid of.

A starting point is that many many people think that our current health care system is a disaster and delivers terrible quality at very high cost.

I would suggest looking at a couple of these items:

1. The book, The Business of Health, by Ohsfeldt and Schneider. Among other things, these two economists show that if you take out the impact of high rates of murder and accidental (especially highway) deaths from the US life expectancy, the US ends up number one in life expectancy. Now how can you possibly say that our high rates of murder are due to failings of the health care system?

2. Read the paper by June and David O'Neill, "Health Status, Health Care and Inequality: Canada vs the US," NBER Working paper13429. They document the failings of comparing health care systems by metrics such as infant mortality and, similar to point number one above, life expectancy. On infant mortality, they show how much of a role low birth weight plays in infant mortality -- and how low birth weight is greatest for teenage mothers. Of course, the US has many more teenage mothers than Canada. Shall we blame teenage pregnancy on the health care system?

3. And the paper by Hall and Jones, "The Value of Life and the Rise in Health Spending," Quarterly Journal of Economics, Feb 2007. These economists argue that optimal health care spending in the US is not inconsistent with utility maximization and that as we grow richer we should expect the percentage of GDP spent on health to rise much higher. This is a great point, and they back it up with great theory. In fact a colleague of mine was recently lamenting how he could not spend as much as he would like on health care, because he cannot hire a doctor on a consulting basis as easily as he would like. This is true: the practices and institutional rules in place do restrict us from buying some health care that many of us would purchase if we could.

4. The paper by my Dartmouth colleague Jon Skinner, and Alan Garber, "Is American Health Care Uniquely Inefficient?," Journal of Economic Perspectives, American Economic Association, vol. 22(4), pages 27-50, Fall. In particular, note the very neat argument how heterogeneity in demand for health care in different regions of a country might make it look like that country is inefficient in producing health care.

And here are some interesting questions to ask folks and ponder their responses:

-- should insurance pay for routine expenses like eye exams, and teeth cleanings?
-- should insurance pay for annual doctor checkups?
-- Why do we assume that people can take their cars in for oil changes, understanding that regular oil changes increase the life of the car, but we cannot trust people to go see the doctor?

Thursday, August 13, 2009

Health Care Incentives

One of my main worries about the health care proposals is that I don't see them doing anything significant to reduce costs -- and even that the whole mindset is misguided on the cost issue. (On this latter point, the correct measure of welfare is clearly not cost, but overall surplus. Even more spending on health care may well be welfare enhancing.)

But if we want to assume that there is spending beyond the point where marginal cost equals marginal benefit, and we want to reduce the total spending, the question is how to do it.

To that end, I would present the case of Dartmouth College. Like many reasonably sized institutions and businesses, the College self-insures in regard to the health expenses of its employees. So while we all think we have Blue Cross Blue Shield insurance, in fact that organization simply manages the administration of the College's health benefits to its employees. The College actually ends up paying all the bills. Of course this would be true even if we did buy true insurance, as over time the premium we would pay would have to equal costs.

My point here is simply that the College, since it bears all of its employees' health costs, has tremendous incentive to reduce those costs through whatever means it can. It can find the most efficient administrative entity to handle all the paperwork -- and it can find this administrator through a competitive process. I have to assume that Anthem BC/BS must offer the best overall package of service and cost. The College can also tell this administrator how to structure the different policies offered to employees, including deductibles, coninsurance, incentives for good behavior, etc. In fact we have a relatively active committee consisting of faculty, staff and administrators who work together to determine the policy options that will be offered each year. I think the overall incentives and ability to manage health care costs are quite high.

Even with these incentives and ability for an institution like Dartmouth to manage health care costs, we still see cost increases very similar to those of the rest of the nation. And I think our overall costs, in an absolute sense, are close to average as well. As a very rough calculation, the cost for a single person to buy health insurance from the College is around $6500. If I take that and multiply by the 300 million population of the US, and divide by US GDP of about 13.8 trillion dollars, I get around 14%, roughly our national expenditures on health care.

So I don't think that the College does any better in reducing costs than the rest of the country.

And this is even with the incentives and ability noted above -- and even with the Dartmouth Institute on campus, the folks who the Obama administration seem to think have the magic wand to show where waste is and how to eliminate it!

So here is the question: to the extent that the health care proposals shift the center of cost controls and incentives away from institutions like Dartmouth and towards the US government, why should we assume there will be better management of health care spending?

Sunday, August 09, 2009

Review of the Health Care Proposals: Predation by Government Enterprises?

It is costly to stay up-to-date on the status of the health care bill specifics, but the broad picture seems pretty well agreed. The Kaiser Foundation provides good summaries, see here. As I read the proposals, I see these as the key features:

1. A mandate for all citizens to have health insurance (punishable with fines).
2. A mandate for all employers (except "small" businesses) to provide health insurance to employees.
3. Provide individuals with subsidies to buy insurance.
4. Provide "small" businesses with subsidies to provide insurance to employees.
5. Create "gateways" or "exchanges" through which individuals and certain employers would buy insurance.

Here are my issues, with the last one being the most serious.

First, if a key objective is to provide health care insurance to the currently uninsured and needy, the proposal will likely fail. The system is too complex. The people that I want to be covered by health insurance are likely to be excluded by this proposal simply because they will still not know what to do or how to get coverage.

Second, the proposal does not cut the tie between health insurance and the employer, in fact it increases the strength of that tie. If you want to increase portability of insurance, get rid of the employer-based nature of our current coverage. That would be a great move in the right direction. Do our employers buy car insurance for us? No. Not even life insurance, at least not most of it. Then why health insurance?

Third, the proposal is economic/social engineering at its worst. The proposal has a bunch of subsidies, the most insidious being those for "small" businesses. Now, I like small business as much as everyone, but I like big business too. Why should our health policy have built into it an advantage for some kind of businesses? Can you imagine how this will be manipulated in years to come? (As an aside, the proposals should be viewed in that light -- as living documents. Don't just take the current policy as the end of the road, but anticipate how it will change with political winds in the future.)

Fourth, and most serious I think, are the "gateways" or "exchanges" that will be created. Think of this gateway idea as a government-overseen consortium of insurers, all of who can offer different health policies to the individuals and employers who qualify to buy insurance through the gateway (a set that even in the current bill is set to expand over time, letting more and more employers and individuals opt into the gateway if they desire). At least in the House bill, there is a provision to ensure that there is a "public" or "community health insurance" option in the gateway. This public option will compete with private insurers. The problem is that the public option is likely to have a huge cost advantage over the private insurers, with the cost advantage arising from the likelihood that the public option will pay service providers (doctors, hospitals, pharma companies) a much lower rate than private insurers can. The version of the House bill that I just read clearly specifies that the public option will pay suppliers prices based off Medicare rates (that are very low compared to what private insurers pay). More on this issue below.

To the extent that the public option has a cost advantage, it is the stealth bomb of the health care proposal. We should expect the public option to be increasingly chosen. As that happens, suppliers will raise prices to the private insurers even more, making their cost disadvantage even worse, and the market share of the public option will increase even more. Sounds to me like a vicious cycle leading to a public monopoly -- hence my reference to predation by government enterprise. Give a government entity a cost advantage, and it will drive private business out.

I recognize there is a vigorous debate over this issue of the public option, and I do have some questions about the theory above. Why can the public option just piggyback on Medicare prices, while private insurers cannot? This is an interesting question. How can Medicare get such low rates today, while private insurers pay more? I suspect there are two factors, one being the sheer size of Medicare and the second being that health providers do not want to turn down Medicare patients for political reasons. These effects will apply just as strongly to the public option described above. Steven Pearlstein, whose editorials I normally find enormously enlightening, has one in the Washington Post today that seems to me to rather shrill and downright wrong. He says that Republicans are "propagating falsehoods" about the proposal, with his main point illustrated with this quote:
But there is no credible way to look at what has been proposed by the president or any congressional committee and conclude that these will result in a government takeover of the health-care system. That is a flat-out lie whose only purpose is to scare the public and stop political conversation.


Now my analysis above might be wrong, and maybe Pearlstein's assurances about House leaders ruling out "piggybacking off Medicare" might turn out correct, but to say that I am being disengenous or being a "political terrorist" by making the argument, well, that is just the kind of discussion around this major policy proposal that we do not need. Lighten up, Pearlstein. You just lost a bunch of credibility from this reader.

I can support a health care proposal that would provide base insurance for the currently uninsured. I suspect many would support that. Why not just create something that did that, and leave the rest alone? And, be honest about what the effect will be: the more people we provide with free health care (free to them), the bigger will be the national bill for health care. This is NOT going to reduce costs.